commit 56b824eb49 upstream.
Currently the skb size after coalescing is only limited by the skb
layout (the skb must not carry frag_list). A single coalesced skb
covering several MSS can potentially fill completely the receive
buffer. In such a case, the snd win will zero until the receive buffer
will be empty again, affecting tput badly.
Fixes: 8268ed4c9d ("mptcp: introduce and use mptcp_try_coalesce()")
Cc: stable@vger.kernel.org # please delay 2 weeks after 6.13-final release
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20241230-net-mptcp-rbuf-fixes-v1-3-8608af434ceb@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 6bb194d036 upstream.
The NCSI state machine as it's currently implemented assumes that
transition to the next logical state is performed either explicitly by
calling `schedule_work(&ndp->work)` to re-queue itself or implicitly
after processing the predefined (ndp->pending_req_num) number of
replies. Thus to avoid the configuration FSM from advancing prematurely
and getting out of sync with the process it's essential to not skip
waiting for a reply.
This patch makes the code wait for reception of the Deselect Package
response for the last package probed before proceeding to channel
configuration.
Thanks go to Potin Lai and Cosmo Chou for the initial investigation and
testing.
Fixes: 8e13f70be0 ("net/ncsi: Probe single packages to avoid conflict")
Cc: stable@vger.kernel.org
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Link: https://patch.msgid.link/20250116152900.8656-1-fercerpav@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 110b43ef05 upstream.
The "pipe" variable is a u8 which comes from the network. If it's more
than 127, then it results in memory corruption in the caller,
nci_hci_connect_gate().
Cc: stable@vger.kernel.org
Fixes: a1b0b94158 ("NFC: nci: Create pipe on specific gate in nci_hci_connect_gate")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patch.msgid.link/bcf5453b-7204-4297-9c20-4d8c7dacf586@stanley.mountain
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 5c61419e02 upstream.
One of the possible ways to enable the input MTU auto-selection for L2CAP
connections is supposed to be through passing a special "0" value for it
as a socket option. Commit [1] added one of those into avdtp. However, it
simply wouldn't work because the kernel still treats the specified value
as invalid and denies the setting attempt. Recorded BlueZ logs include the
following:
bluetoothd[496]: profiles/audio/avdtp.c:l2cap_connect() setsockopt(L2CAP_OPTIONS): Invalid argument (22)
[1]: ae5be371a9
Found by Linux Verification Center (linuxtesting.org).
Fixes: 4b6e228e29 ("Bluetooth: Auto tune if input MTU is set to 0")
Cc: stable@vger.kernel.org
Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 5f397409f8 upstream.
A NULL sock pointer is passed into l2cap_sock_alloc() when it is called
from l2cap_sock_new_connection_cb() and the error handling paths should
also be aware of it.
Seemingly a more elegant solution would be to swap bt_sock_alloc() and
l2cap_chan_create() calls since they are not interdependent to that moment
but then l2cap_chan_create() adds the soon to be deallocated and still
dummy-initialized channel to the global list accessible by many L2CAP
paths. The channel would be removed from the list in short period of time
but be a bit more straight-forward here and just check for NULL instead of
changing the order of function calls.
Found by Linux Verification Center (linuxtesting.org) with SVACE static
analysis tool.
Fixes: 7c4f78cdb8 ("Bluetooth: L2CAP: do not leave dangling sk pointer on error in l2cap_sock_create()")
Cc: stable@vger.kernel.org
Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 41b996ce83 ]
The RXRPC_CALL_SERVER_SECURING state doesn't really belong with the other
states in the call's state set as the other states govern the call's Rx/Tx
phase transition and govern when packets can and can't be received or
transmitted. The "Securing" state doesn't actually govern the reception of
packets and would need to be split depending on whether or not we've
received the last packet yet (to mirror RECV_REQUEST/ACK_REQUEST).
The "Securing" state is more about whether or not we can start forwarding
packets to the application as recvmsg will need to decode them and the
decoding can't take place until the challenge/response exchange has
completed.
Fix this by removing the RXRPC_CALL_SERVER_SECURING state from the state
set and, instead, using a flag, RXRPC_CALL_CONN_CHALLENGING, to track
whether or not we can queue the call for reception by recvmsg() or notify
the kernel app that data is ready. In the event that we've already
received all the packets, the connection event handler will poke the app
layer in the appropriate manner.
Also there's a race whereby the app layer sees the last packet before rxrpc
has managed to end the rx phase and change the state to one amenable to
allowing a reply. Fix this by queuing the packet after calling
rxrpc_end_rx_phase().
Fixes: 17926a7932 ("[AF_RXRPC]: Provide secure RxRPC sockets for use by userspace and kernel both")
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Simon Horman <horms@kernel.org>
cc: linux-afs@lists.infradead.org
Link: https://patch.msgid.link/20250204230558.712536-2-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 638ba50893 ]
qdisc_tree_reduce_backlog() notifies parent qdisc only if child
qdisc becomes empty, therefore we need to reduce the backlog of the
child qdisc before calling it. Otherwise it would miss the opportunity
to call cops->qlen_notify(), in the case of DRR, it resulted in UAF
since DRR uses ->qlen_notify() to maintain its active list.
Fixes: f8d4bc4550 ("net/sched: netem: account for backlog updates from child qdisc")
Cc: Martin Ottens <martin.ottens@fau.de>
Reported-by: Mingi Cho <mincho@theori.io>
Signed-off-by: Cong Wang <cong.wang@bytedance.com>
Link: https://patch.msgid.link/20250204005841.223511-4-xiyou.wangcong@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 647cef20e6 ]
Expected behaviour:
In case we reach scheduler's limit, pfifo_tail_enqueue() will drop a
packet in scheduler's queue and decrease scheduler's qlen by one.
Then, pfifo_tail_enqueue() enqueue new packet and increase
scheduler's qlen by one. Finally, pfifo_tail_enqueue() return
`NET_XMIT_CN` status code.
Weird behaviour:
In case we set `sch->limit == 0` and trigger pfifo_tail_enqueue() on a
scheduler that has no packet, the 'drop a packet' step will do nothing.
This means the scheduler's qlen still has value equal 0.
Then, we continue to enqueue new packet and increase scheduler's qlen by
one. In summary, we can leverage pfifo_tail_enqueue() to increase qlen by
one and return `NET_XMIT_CN` status code.
The problem is:
Let's say we have two qdiscs: Qdisc_A and Qdisc_B.
- Qdisc_A's type must have '->graft()' function to create parent/child relationship.
Let's say Qdisc_A's type is `hfsc`. Enqueue packet to this qdisc will trigger `hfsc_enqueue`.
- Qdisc_B's type is pfifo_head_drop. Enqueue packet to this qdisc will trigger `pfifo_tail_enqueue`.
- Qdisc_B is configured to have `sch->limit == 0`.
- Qdisc_A is configured to route the enqueued's packet to Qdisc_B.
Enqueue packet through Qdisc_A will lead to:
- hfsc_enqueue(Qdisc_A) -> pfifo_tail_enqueue(Qdisc_B)
- Qdisc_B->q.qlen += 1
- pfifo_tail_enqueue() return `NET_XMIT_CN`
- hfsc_enqueue() check for `NET_XMIT_SUCCESS` and see `NET_XMIT_CN` => hfsc_enqueue() don't increase qlen of Qdisc_A.
The whole process lead to a situation where Qdisc_A->q.qlen == 0 and Qdisc_B->q.qlen == 1.
Replace 'hfsc' with other type (for example: 'drr') still lead to the same problem.
This violate the design where parent's qlen should equal to the sum of its childrens'qlen.
Bug impact: This issue can be used for user->kernel privilege escalation when it is reachable.
Fixes: 57dbb2d83d ("sched: add head drop fifo queue")
Reported-by: Quang Le <quanglex97@gmail.com>
Signed-off-by: Quang Le <quanglex97@gmail.com>
Signed-off-by: Cong Wang <cong.wang@bytedance.com>
Link: https://patch.msgid.link/20250204005841.223511-2-xiyou.wangcong@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 4241a702e0 ]
The rxrpc_connection attend queue is never used because conn::attend_link
is never initialised and so is always NULL'd out and thus always appears to
be busy. This requires the following fix:
(1) Fix this the attend queue problem by initialising conn::attend_link.
And, consequently, two further fixes for things masked by the above bug:
(2) Fix rxrpc_input_conn_event() to handle being invoked with a NULL
sk_buff pointer - something that can now happen with the above change.
(3) Fix the RXRPC_SKB_MARK_SERVICE_CONN_SECURED message to carry a pointer
to the connection and a ref on it.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Jakub Kicinski <kuba@kernel.org>
cc: "David S. Miller" <davem@davemloft.net>
cc: Eric Dumazet <edumazet@google.com>
cc: Paolo Abeni <pabeni@redhat.com>
cc: Simon Horman <horms@kernel.org>
cc: linux-afs@lists.infradead.org
cc: netdev@vger.kernel.org
Fixes: f2cce89a07 ("rxrpc: Implement a mechanism to send an event notification to a connection")
Link: https://patch.msgid.link/20250203110307.7265-3-dhowells@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 244f8aa46f ]
Commit ec6e57beaf ("ethtool: rss: don't report key if device
doesn't support it") intended to stop reporting key fields for
additional rss contexts if device has a global hashing key.
Later we added dump support and the filtering wasn't properly
added there. So we end up reporting the key fields in dumps
but not in dos:
# ./pyynl/cli.py --spec netlink/specs/ethtool.yaml --do rss-get \
--json '{"header": {"dev-index":2}, "context": 1 }'
{
"header": { ... },
"context": 1,
"indir": [0, 1, 2, 3, ...]]
}
# ./pyynl/cli.py --spec netlink/specs/ethtool.yaml --dump rss-get
[
... snip context 0 ...
{ "header": { ... },
"context": 1,
"indir": [0, 1, 2, 3, ...],
-> "input_xfrm": 255,
-> "hfunc": 1,
-> "hkey": "000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
}
]
Hide these fields correctly.
The drivers/net/hw/rss_ctx.py selftest catches this when run on
a device with single key, already:
# Check| At /root/./ksft-net-drv/drivers/net/hw/rss_ctx.py, line 381, in test_rss_context_dump:
# Check| ksft_ne(set(data.get('hkey', [1])), {0}, "key is all zero")
# Check failed {0} == {0} key is all zero
not ok 8 rss_ctx.test_rss_context_dump
Fixes: f6122900f4 ("ethtool: rss: support dumping RSS contexts")
Reviewed-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Joe Damato <jdamato@fastly.com>
Link: https://patch.msgid.link/20250201013040.725123-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 235174b2be ]
Commit 4094871db1 ("udp: only do GSO if # of segs > 1") avoided GSO
for small packets. But the kernel currently dismisses GSO requests only
after checking MTU/PMTU on gso_size. This means any packets, regardless
of their payload sizes, could be dropped when PMTU becomes smaller than
requested gso_size. We encountered this issue in production and it
caused a reliability problem that new QUIC connection cannot be
established before PMTU cache expired, while non GSO sockets still
worked fine at the same time.
Ideally, do not check any GSO related constraints when payload size is
smaller than requested gso_size, and return EMSGSIZE instead of EINVAL
on MTU/PMTU check failure to be more specific on the error cause.
Fixes: 4094871db1 ("udp: only do GSO if # of segs > 1")
Signed-off-by: Yan Zhai <yan@cloudflare.com>
Suggested-by: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 5fe71fda89 ]
On a 32bit system the "keylen + sizeof(struct tipc_aead_key)" math could
have an integer wrapping issue. It doesn't matter because the "keylen"
is checked on the next line, but just to make life easier for static
analysis tools, let's re-order these conditions and avoid the integer
overflow.
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 2b91cc1214 upstream.
The info.flow_type is for RXFH commands, ntuple flow_type is inside
the flow spec. The check currently does nothing, as info.flow_type
is 0 (or even uninitialized by user space) for ETHTOOL_SRXCLSRLINS.
Fixes: 9e43ad7a1e ("net: ethtool: only allow set_rxnfc with rss + ring_cookie if driver opts in")
Reviewed-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Joe Damato <jdamato@fastly.com>
Link: https://patch.msgid.link/20250201013040.725123-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 9407190947 upstream.
The check for non-zero ring with RSS is only relevant for
ETHTOOL_SRXCLSRLINS command, in other cases the check tries to access
memory which was not initialized by the userspace tool. Only perform the
check in case of ETHTOOL_SRXCLSRLINS.
Without this patch, filter deletion (for example) could statistically
result in a false error:
# ethtool --config-ntuple eth3 delete 484
rmgr: Cannot delete RX class rule: Invalid argument
Cannot delete classification rule
Fixes: 9e43ad7a1e ("net: ethtool: only allow set_rxnfc with rss + ring_cookie if driver opts in")
Link: https://lore.kernel.org/netdev/871a9ecf-1e14-40dd-bbd7-e90c92f89d47@nvidia.com/
Reviewed-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Edward Cree <ecree.xilinx@gmail.com>
Link: https://patch.msgid.link/20241202164805.1637093-1-gal@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit a35672819f upstream.
A recent commit jumped over the dst hash computation and
left the symbol uninitialized. Fix this by explicitly
computing the dst hash before it is used.
Fixes: 0045e3d806 ("xfrm: Cache used outbound xfrm states at the policy.")
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 9d287e70c5 upstream.
Error handling is missing when call to nla_put_u32() fails.
Handle the error when the call to nla_put_u32() returns an error.
The error was reported by Coverity Scan.
Report:
CID 1601525: (#1 of 1): Unused value (UNUSED_VALUE)
returned_value: Assigning value from nla_put_u32(skb, XFRMA_SA_PCPU, x->pcpu_num)
to err here, but that stored value is overwritten before it can be used
Fixes: 1ddf9916ac ("xfrm: Add support for per cpu xfrm state handling.")
Signed-off-by: Everest K.C. <everestkc@everestkc.com.np>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit e598d8981f upstream.
The Fixes commit mentioned this:
> An MPTCP firewall blackhole can be detected if the following SYN
> retransmission after a fallback to "plain" TCP is accepted.
But in fact, this blackhole was detected if any following SYN
retransmissions after a fallback to TCP was accepted.
That's because 'mptcp_subflow_early_fallback()' will set 'request_mptcp'
to 0, and 'mpc_drop' will never be reset to 0 after.
This is an issue, because some not so unusual situations might cause the
kernel to detect a false-positive blackhole, e.g. a client trying to
connect to a server while the network is not ready yet, causing a few
SYN retransmissions, before reaching the end server.
Fixes: 27069e7cb3 ("mptcp: disable active MPTCP in case of blackhole")
Cc: stable@vger.kernel.org
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 1b9335a800 upstream.
The field length description provides the length of each separated key
field in the concatenation, each field gets rounded up to 32-bits to
calculate the pipapo rule width from pipapo_init(). The set key length
provides the total size of the key aligned to 32-bits.
Register-based arithmetics still allows for combining mismatching set
key length and field length description, eg. set key length 10 and field
description [ 5, 4 ] leading to pipapo width of 12.
Cc: stable@vger.kernel.org
Fixes: 3ce67e3793 ("netfilter: nf_tables: do not allow mismatch field size and set key length")
Reported-by: Noam Rathaus <noamr@ssd-disclosure.com>
Reviewed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 966a675da8 upstream.
I noticed that a handful of NFSv3 fstests were taking an
unexpectedly long time to run. Troubleshooting showed that the
server's TCP window closed and never re-opened, which caused the
client to trigger an RPC retransmit timeout after 180 seconds.
The client's recovery action was to establish a fresh connection
and retransmit the timed-out requests. This worked, but it adds a
long delay.
I tracked the problem to the commit that attempted to reduce the
rate at which the network layer delivers TCP socket data_ready
callbacks. Under most circumstances this change worked as expected,
but for NFSv3, which has no session or other type of throttling, it
can overwhelm the receiver on occasion.
I'm sure I could tweak the lowat settings, but the small benefit
doesn't seem worth the bother. Just revert it.
Fixes: 2b877fc53e ("SUNRPC: Reduce thread wake-up rate when receiving large RPC messages")
Cc: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 3595599fa8 ]
Device-bound programs are used to support RX metadata kfuncs. These
kfuncs are driver-specific and rely on the driver context to read the
metadata. This means they can't work in generic XDP mode. However, there
is no check to disallow such programs from being attached in generic
mode, in which case the metadata kfuncs will be called in an invalid
context, leading to crashes.
Fix this by adding a check to disallow attaching device-bound programs
in generic mode.
Fixes: 2b3486bc2d ("bpf: Introduce device-bound XDP programs")
Reported-by: Marcus Wichelmann <marcus.wichelmann@hetzner-cloud.de>
Closes: https://lore.kernel.org/r/dae862ec-43b5-41a0-8edf-46c59071cdda@hetzner-cloud.de
Tested-by: Marcus Wichelmann <marcus.wichelmann@hetzner-cloud.de>
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/20250127131344.238147-1-toke@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 8c670bdfa5 ]
Testing with iperf3 using the "pasta" protocol splicer has revealed
a problem in the way tcp handles window advertising in extreme memory
squeeze situations.
Under memory pressure, a socket endpoint may temporarily advertise
a zero-sized window, but this is not stored as part of the socket data.
The reasoning behind this is that it is considered a temporary setting
which shouldn't influence any further calculations.
However, if we happen to stall at an unfortunate value of the current
window size, the algorithm selecting a new value will consistently fail
to advertise a non-zero window once we have freed up enough memory.
This means that this side's notion of the current window size is
different from the one last advertised to the peer, causing the latter
to not send any data to resolve the sitution.
The problem occurs on the iperf3 server side, and the socket in question
is a completely regular socket with the default settings for the
fedora40 kernel. We do not use SO_PEEK or SO_RCVBUF on the socket.
The following excerpt of a logging session, with own comments added,
shows more in detail what is happening:
// tcp_v4_rcv(->)
// tcp_rcv_established(->)
[5201<->39222]: ==== Activating log @ net/ipv4/tcp_input.c/tcp_data_queue()/5257 ====
[5201<->39222]: tcp_data_queue(->)
[5201<->39222]: DROPPING skb [265600160..265665640], reason: SKB_DROP_REASON_PROTO_MEM
[rcv_nxt 265600160, rcv_wnd 262144, snt_ack 265469200, win_now 131184]
[copied_seq 259909392->260034360 (124968), unread 5565800, qlen 85, ofoq 0]
[OFO queue: gap: 65480, len: 0]
[5201<->39222]: tcp_data_queue(<-)
[5201<->39222]: __tcp_transmit_skb(->)
[tp->rcv_wup: 265469200, tp->rcv_wnd: 262144, tp->rcv_nxt 265600160]
[5201<->39222]: tcp_select_window(->)
[5201<->39222]: (inet_csk(sk)->icsk_ack.pending & ICSK_ACK_NOMEM) ? --> TRUE
[tp->rcv_wup: 265469200, tp->rcv_wnd: 262144, tp->rcv_nxt 265600160]
returning 0
[5201<->39222]: tcp_select_window(<-)
[5201<->39222]: ADVERTISING WIN 0, ACK_SEQ: 265600160
[5201<->39222]: [__tcp_transmit_skb(<-)
[5201<->39222]: tcp_rcv_established(<-)
[5201<->39222]: tcp_v4_rcv(<-)
// Receive queue is at 85 buffers and we are out of memory.
// We drop the incoming buffer, although it is in sequence, and decide
// to send an advertisement with a window of zero.
// We don't update tp->rcv_wnd and tp->rcv_wup accordingly, which means
// we unconditionally shrink the window.
[5201<->39222]: tcp_recvmsg_locked(->)
[5201<->39222]: __tcp_cleanup_rbuf(->) tp->rcv_wup: 265469200, tp->rcv_wnd: 262144, tp->rcv_nxt 265600160
[5201<->39222]: [new_win = 0, win_now = 131184, 2 * win_now = 262368]
[5201<->39222]: [new_win >= (2 * win_now) ? --> time_to_ack = 0]
[5201<->39222]: NOT calling tcp_send_ack()
[tp->rcv_wup: 265469200, tp->rcv_wnd: 262144, tp->rcv_nxt 265600160]
[5201<->39222]: __tcp_cleanup_rbuf(<-)
[rcv_nxt 265600160, rcv_wnd 262144, snt_ack 265469200, win_now 131184]
[copied_seq 260040464->260040464 (0), unread 5559696, qlen 85, ofoq 0]
returning 6104 bytes
[5201<->39222]: tcp_recvmsg_locked(<-)
// After each read, the algorithm for calculating the new receive
// window in __tcp_cleanup_rbuf() finds it is too small to advertise
// or to update tp->rcv_wnd.
// Meanwhile, the peer thinks the window is zero, and will not send
// any more data to trigger an update from the interrupt mode side.
[5201<->39222]: tcp_recvmsg_locked(->)
[5201<->39222]: __tcp_cleanup_rbuf(->) tp->rcv_wup: 265469200, tp->rcv_wnd: 262144, tp->rcv_nxt 265600160
[5201<->39222]: [new_win = 262144, win_now = 131184, 2 * win_now = 262368]
[5201<->39222]: [new_win >= (2 * win_now) ? --> time_to_ack = 0]
[5201<->39222]: NOT calling tcp_send_ack()
[tp->rcv_wup: 265469200, tp->rcv_wnd: 262144, tp->rcv_nxt 265600160]
[5201<->39222]: __tcp_cleanup_rbuf(<-)
[rcv_nxt 265600160, rcv_wnd 262144, snt_ack 265469200, win_now 131184]
[copied_seq 260099840->260171536 (71696), unread 5428624, qlen 83, ofoq 0]
returning 131072 bytes
[5201<->39222]: tcp_recvmsg_locked(<-)
// The above pattern repeats again and again, since nothing changes
// between the reads.
[...]
[5201<->39222]: tcp_recvmsg_locked(->)
[5201<->39222]: __tcp_cleanup_rbuf(->) tp->rcv_wup: 265469200, tp->rcv_wnd: 262144, tp->rcv_nxt 265600160
[5201<->39222]: [new_win = 262144, win_now = 131184, 2 * win_now = 262368]
[5201<->39222]: [new_win >= (2 * win_now) ? --> time_to_ack = 0]
[5201<->39222]: NOT calling tcp_send_ack()
[tp->rcv_wup: 265469200, tp->rcv_wnd: 262144, tp->rcv_nxt 265600160]
[5201<->39222]: __tcp_cleanup_rbuf(<-)
[rcv_nxt 265600160, rcv_wnd 262144, snt_ack 265469200, win_now 131184]
[copied_seq 265600160->265600160 (0), unread 0, qlen 0, ofoq 0]
returning 54672 bytes
[5201<->39222]: tcp_recvmsg_locked(<-)
// The receive queue is empty, but no new advertisement has been sent.
// The peer still thinks the receive window is zero, and sends nothing.
// We have ended up in a deadlock situation.
Note that well behaved endpoints will send win0 probes, so the problem
will not occur.
Furthermore, we have observed that in these situations this side may
send out an updated 'th->ack_seq´ which is not stored in tp->rcv_wup
as it should be. Backing ack_seq seems to be harmless, but is of
course still wrong from a protocol viewpoint.
We fix this by updating the socket state correctly when a packet has
been dropped because of memory exhaustion and we have to advertize
a zero window.
Further testing shows that the connection recovers neatly from the
squeeze situation, and traffic can continue indefinitely.
Fixes: e2142825c1 ("net: tcp: send zero-window ACK when no memory")
Cc: Menglong Dong <menglong8.dong@gmail.com>
Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Link: https://patch.msgid.link/20250127231304.1465565-1-jmaloy@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit aa388c7211 ]
sk_err is set when a (connectible) connect() fails. Effectively, this makes
an otherwise still healthy SS_UNCONNECTED socket impossible to use for any
subsequent connection attempts.
Clear sk_err upon trying to establish a connection.
Fixes: d021c34405 ("VSOCK: Introduce VM Sockets")
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Reviewed-by: Luigi Leonardi <leonardi@redhat.com>
Signed-off-by: Michal Luczaj <mhal@rbox.co>
Link: https://patch.msgid.link/20250128-vsock-transport-vs-autobind-v3-2-1cf57065b770@rbox.co
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 4f5a52adeb ]
The sanity check that both source and destination are set when symmetric
RSS hash is requested is only relevant for ETHTOOL_SRXFH (rx-flow-hash),
it should not be performed on any other commands (e.g.
ETHTOOL_SRXCLSRLINS/ETHTOOL_SRXCLSRLDEL).
This resolves accessing uninitialized 'info.data' field, and fixes false
errors in rule insertion:
# ethtool --config-ntuple eth2 flow-type ip4 dst-ip 255.255.255.255 action -1 loc 0
rmgr: Cannot insert RX class rule: Invalid argument
Cannot insert classification rule
Fixes: 13e59344fb ("net: ethtool: add support for symmetric-xor RSS hash")
Cc: Ahmed Zaki <ahmed.zaki@intel.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Reviewed-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Ahmed Zaki <ahmed.zaki@intel.com>
Link: https://patch.msgid.link/20250126191845.316589-1-gal@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 9e43ad7a1e ]
Ethtool ntuple filters with FLOW_RSS were originally defined as adding
the base queue ID (ring_cookie) to the value from the indirection table,
so that the same table could distribute over more than one set of queues
when used by different filters.
However, some drivers / hardware ignore the ring_cookie, and simply use
the indirection table entries as queue IDs directly. Thus, for drivers
which have not opted in by setting ethtool_ops.cap_rss_rxnfc_adds to
declare that they support the original (addition) semantics, reject in
ethtool_set_rxnfc any filter which combines FLOW_RSS and a nonzero ring.
(For a ring_cookie of zero, both behaviours are equivalent.)
Set the cap bit in sfc, as it is known to support this feature.
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Martin Habets <habetsm.xilinx@gmail.com>
Link: https://patch.msgid.link/cc3da0844083b0e301a33092a6299e4042b65221.1731499022.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Stable-dep-of: 4f5a52adeb ("ethtool: Fix set RXNFC command with symmetric RSS hash")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 79d458c130 ]
In its address list, afs now retains pointers to and refs on one or more
rxrpc_peer objects. The address list is freed under RCU and at this time,
it puts the refs on those peers.
Now, when an rxrpc_peer object runs out of refs, it gets removed from the
peer hash table and, for that, rxrpc has to take a spinlock. However, it
is now being called from afs's RCU cleanup, which takes place in BH
context - but it is just taking an ordinary spinlock.
The put may also be called from non-BH context, and so there exists the
possibility of deadlock if the BH-based RCU cleanup happens whilst the hash
spinlock is held. This led to the attached lockdep complaint.
Fix this by changing spinlocks of rxnet->peer_hash_lock back to
BH-disabling locks.
================================
WARNING: inconsistent lock state
6.13.0-rc5-build2+ #1223 Tainted: G E
--------------------------------
inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage.
swapper/1/0 [HC0[0]:SC1[1]:HE1:SE0] takes:
ffff88810babe228 (&rxnet->peer_hash_lock){+.?.}-{3:3}, at: rxrpc_put_peer+0xcb/0x180
{SOFTIRQ-ON-W} state was registered at:
mark_usage+0x164/0x180
__lock_acquire+0x544/0x990
lock_acquire.part.0+0x103/0x280
_raw_spin_lock+0x2f/0x40
rxrpc_peer_keepalive_worker+0x144/0x440
process_one_work+0x486/0x7c0
process_scheduled_works+0x73/0x90
worker_thread+0x1c8/0x2a0
kthread+0x19b/0x1b0
ret_from_fork+0x24/0x40
ret_from_fork_asm+0x1a/0x30
irq event stamp: 972402
hardirqs last enabled at (972402): [<ffffffff8244360e>] _raw_spin_unlock_irqrestore+0x2e/0x50
hardirqs last disabled at (972401): [<ffffffff82443328>] _raw_spin_lock_irqsave+0x18/0x60
softirqs last enabled at (972300): [<ffffffff810ffbbe>] handle_softirqs+0x3ee/0x430
softirqs last disabled at (972313): [<ffffffff810ffc54>] __irq_exit_rcu+0x44/0x110
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(&rxnet->peer_hash_lock);
<Interrupt>
lock(&rxnet->peer_hash_lock);
*** DEADLOCK ***
1 lock held by swapper/1/0:
#0: ffffffff83576be0 (rcu_callback){....}-{0:0}, at: rcu_lock_acquire+0x7/0x30
stack backtrace:
CPU: 1 UID: 0 PID: 0 Comm: swapper/1 Tainted: G E 6.13.0-rc5-build2+ #1223
Tainted: [E]=UNSIGNED_MODULE
Hardware name: ASUS All Series/H97-PLUS, BIOS 2306 10/09/2014
Call Trace:
<IRQ>
dump_stack_lvl+0x57/0x80
print_usage_bug.part.0+0x227/0x240
valid_state+0x53/0x70
mark_lock_irq+0xa5/0x2f0
mark_lock+0xf7/0x170
mark_usage+0xe1/0x180
__lock_acquire+0x544/0x990
lock_acquire.part.0+0x103/0x280
_raw_spin_lock+0x2f/0x40
rxrpc_put_peer+0xcb/0x180
afs_free_addrlist+0x46/0x90 [kafs]
rcu_do_batch+0x2d2/0x640
rcu_core+0x2f7/0x350
handle_softirqs+0x1ee/0x430
__irq_exit_rcu+0x44/0x110
irq_exit_rcu+0xa/0x30
sysvec_apic_timer_interrupt+0x7f/0xa0
</IRQ>
Fixes: 72904d7b9b ("rxrpc, afs: Allow afs to pin rxrpc_peer objects")
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Simon Horman <horms@kernel.org>
cc: linux-afs@lists.infradead.org
Link: https://patch.msgid.link/2095618.1737622752@warthog.procyon.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 05d91cdb1f ]
Copy of the rationale from 790071347a:
Change ndo_set_mac_address to dev_set_mac_address because
dev_set_mac_address provides a way to notify network layer about MAC
change. In other case, services may not aware about MAC change and keep
using old one which set from network adapter driver.
As example, DHCP client from systemd do not update MAC address without
notification from net subsystem which leads to the problem with acquiring
the right address from DHCP server.
Since dev_set_mac_address requires RTNL lock the operation can not be
performed directly in the response handler, see
9e2bbab94b.
The way of selecting the first suitable MAC address from the list is
changed, instead of having the driver check it this patch just assumes
any valid MAC should be good.
Fixes: b8291cf3d1 ("net/ncsi: Add NC-SI 1.2 Get MC MAC Address command")
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 6c9b7db96d ]
For the state cache lookup xfrm_input_state_lookup() first disables
preemption, to remain on the CPU and then retrieves a per-CPU pointer.
Within the preempt-disable section it also acquires
netns_xfrm::xfrm_state_lock, a spinlock_t. This lock must not be
acquired with explicit disabled preemption (such as by get_cpu())
because this lock becomes a sleeping lock on PREEMPT_RT.
To remain on the same CPU is just an optimisation for the CPU local
lookup. The actual modification of the per-CPU variable happens with
netns_xfrm::xfrm_state_lock acquired.
Remove get_cpu() and use the state_cache_input on the current CPU.
Reported-by: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Closes: https://lore.kernel.org/all/CAADnVQKkCLaj=roayH=Mjiiqz_svdf1tsC3OE4EC0E=mAD+L1A@mail.gmail.com/
Fixes: 81a331a0e7 ("xfrm: Add an inbound percpu state cache.")
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e952837f3d ]
lookup and resize can run in parallel.
The xfrm_state_hash_generation seqlock ensures a retry, but the hash
functions can observe a hmask value that is too large for the new hlist
array.
rehash does:
rcu_assign_pointer(net->xfrm.state_bydst, ndst) [..]
net->xfrm.state_hmask = nhashmask;
While state lookup does:
h = xfrm_dst_hash(net, daddr, saddr, tmpl->reqid, encap_family);
hlist_for_each_entry_rcu(x, net->xfrm.state_bydst + h, bydst) {
This is only safe in case the update to state_bydst is larger than
net->xfrm.xfrm_state_hmask (or if the lookup function gets
serialized via state spinlock again).
Fix this by prefetching state_hmask and the associated pointers.
The xfrm_state_hash_generation seqlock retry will ensure that the pointer
and the hmask will be consistent.
The existing helpers, like xfrm_dst_hash(), are now unsafe for RCU side,
add lockdep assertions to document that they are only safe for insert
side.
xfrm_state_lookup_byaddr() uses the spinlock rather than RCU.
AFAICS this is an oversight from back when state lookup was converted to
RCU, this lock should be replaced with RCU in a future patch.
Reported-by: syzbot+5f9f31cb7d985f584d8e@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/netdev/CACT4Y+azwfrE3uz6A5ZErov5YN2LYBN5KrsymBerT36VU8qzBA@mail.gmail.com/
Diagnosed-by: Dmitry Vyukov <dvyukov@google.com>
Fixes: c2f672fc94 ("xfrm: state lookup can be lockless")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 81a331a0e7 ]
Now that we can have percpu xfrm states, the number of active
states might increase. To get a better lookup performance,
we add a percpu cache to cache the used inbound xfrm states.
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Tested-by: Antony Antony <antony.antony@secunet.com>
Tested-by: Tobias Brunner <tobias@strongswan.org>
Stable-dep-of: e952837f3d ("xfrm: state: fix out-of-bounds read during lookup")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 0045e3d806 ]
Now that we can have percpu xfrm states, the number of active
states might increase. To get a better lookup performance,
we cache the used xfrm states at the policy for outbound
IPsec traffic.
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Tested-by: Antony Antony <antony.antony@secunet.com>
Tested-by: Tobias Brunner <tobias@strongswan.org>
Stable-dep-of: e952837f3d ("xfrm: state: fix out-of-bounds read during lookup")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 1ddf9916ac ]
Currently all flows for a certain SA must be processed by the same
cpu to avoid packet reordering and lock contention of the xfrm
state lock.
To get rid of this limitation, the IETF standardized per cpu SAs
in RFC 9611. This patch implements the xfrm part of it.
We add the cpu as a lookup key for xfrm states and a config option
to generate acquire messages for each cpu.
With that, we can have on each cpu a SA with identical traffic selector
so that flows can be processed in parallel on all cpus.
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Tested-by: Antony Antony <antony.antony@secunet.com>
Tested-by: Tobias Brunner <tobias@strongswan.org>
Stable-dep-of: e952837f3d ("xfrm: state: fix out-of-bounds read during lookup")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit c05c5e5aa1 ]
When skb needs GSO and wrap around happens, if xo->seq.low (seqno of
the first skb segment) is before the last seq number but oseq (seqno
of the last segment) is after it, xo->seq.low is still bigger than
replay_esn->oseq while oseq is smaller than it, so the update of
replay_esn->oseq_hi is missed for this case wrap around because of
the change in the cited commit.
For example, if sending a packet with gso_segs=3 while old
replay_esn->oseq=0xfffffffe, we calculate:
xo->seq.low = 0xfffffffe + 1 = 0x0xffffffff
oseq = 0xfffffffe + 3 = 0x1
(oseq < replay_esn->oseq) is true, but (xo->seq.low <
replay_esn->oseq) is false, so replay_esn->oseq_hi is not incremented.
To fix this issue, change the outer checking back for the update of
replay_esn->oseq_hi. And add new checking inside for the update of
packet's oseq_hi.
Fixes: 4b549ccce9 ("xfrm: replay: Fix ESN wrap around for GSO")
Signed-off-by: Jianbo Liu <jianbol@nvidia.com>
Reviewed-by: Patrisious Haddad <phaddad@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 8ac412a336 ]
MEM_WRITE attribute is defined as: "Non-presence of MEM_WRITE means that
MEM is only being read". bpf_load_hdr_opt() both reads and writes from
its arg2 - void *search_res.
This matters a lot for the next commit where we more precisely track
stack accesses. Without this annotation, the verifier will make false
assumptions about the contents of memory written to by helpers and
possibly prune valid branches.
Fixes: 6fad274f06 ("bpf: Add MEM_WRITE attribute")
Acked-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/730e45f8c39be2a5f3d8c4406cceca9d574cbf14.1736886479.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit d640627663 ]
In case of possible unpredictably large arguments passed to
rose_setsockopt() and multiplied by extra values on top of that,
integer overflows may occur.
Do the safest minimum and fix these issues by checking the
contents of 'opt' and returning -EINVAL if they are too large. Also,
switch to unsigned int and remove useless check for negative 'opt'
in ROSE_IDLE case.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Nikita Zhandarovich <n.zhandarovich@fintech.ru>
Link: https://patch.msgid.link/20250115164220.19954-1-n.zhandarovich@fintech.ru
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 25c1a9ca53 ]
I noticed that HyStart incorrectly marks the start of rounds,
leading to inaccurate measurements of ACK train lengths and
resetting the `ca->sample_cnt` variable. This inaccuracy can impact
HyStart's functionality in terminating exponential cwnd growth during
Slow-Start, potentially degrading TCP performance.
The issue arises because the changes introduced in commit 4e1fddc98d
("tcp_cubic: fix spurious Hystart ACK train detections for not-cwnd-limited flows")
moved the caller of the `bictcp_hystart_reset` function inside the `hystart_update` function.
This modification added an additional condition for triggering the caller,
requiring that (tcp_snd_cwnd(tp) >= hystart_low_window) must also
be satisfied before invoking `bictcp_hystart_reset`.
This fix ensures that `bictcp_hystart_reset` is correctly called
at the start of a new round, regardless of the congestion window size.
This is achieved by moving the condition
(tcp_snd_cwnd(tp) >= hystart_low_window)
from before calling `bictcp_hystart_reset` to after it.
I tested with a client and a server connected through two Linux software routers.
In this setup, the minimum RTT was 150 ms, the bottleneck bandwidth was 50 Mbps,
and the bottleneck buffer size was 1 BDP, calculated as (50M / 1514 / 8) * 0.150 = 619 packets.
I conducted the test twice, transferring data from the server to the client for 1.5 seconds.
Before the patch was applied, HYSTART-DELAY stopped the exponential growth of cwnd when
cwnd = 516, and the bottleneck link was not yet saturated (516 < 619).
After the patch was applied, HYSTART-ACK-TRAIN stopped the exponential growth of cwnd when
cwnd = 632, and the bottleneck link was saturated (632 > 619).
In this test, applying the patch resulted in 300 KB more data delivered.
Fixes: 4e1fddc98d ("tcp_cubic: fix spurious Hystart ACK train detections for not-cwnd-limited flows")
Signed-off-by: Mahdi Arghavani <ma.arghavani@yahoo.com>
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Haibo Zhang <haibo.zhang@otago.ac.nz>
Cc: David Eyers <david.eyers@otago.ac.nz>
Cc: Abbas Arghavani <abbas.arghavani@mdu.se>
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Tested-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit a12c76a033 ]
This patch addresses issues with filter counting in block (tcf_block),
particularly for software bypass scenarios, by introducing a more
accurate mechanism using useswcnt.
Previously, filtercnt and skipswcnt were introduced by:
Commit 2081fd3445 ("net: sched: cls_api: add filter counter") and
Commit f631ef39d8 ("net: sched: cls_api: add skip_sw counter")
filtercnt tracked all tp (tcf_proto) objects added to a block, and
skipswcnt counted tp objects with the skipsw attribute set.
The problem is: a single tp can contain multiple filters, some with skipsw
and others without. The current implementation fails in the case:
When the first filter in a tp has skipsw, both skipswcnt and filtercnt
are incremented, then adding a second filter without skipsw to the same
tp does not modify these counters because tp->counted is already set.
This results in bypass software behavior based solely on skipswcnt
equaling filtercnt, even when the block includes filters without
skipsw. Consequently, filters without skipsw are inadvertently bypassed.
To address this, the patch introduces useswcnt in block to explicitly count
tp objects containing at least one filter without skipsw. Key changes
include:
Whenever a filter without skipsw is added, its tp is marked with usesw
and counted in useswcnt. tc_run() now uses useswcnt to determine software
bypass, eliminating reliance on filtercnt and skipswcnt.
This refined approach prevents software bypass for blocks containing
mixed filters, ensuring correct behavior in tc_run().
Additionally, as atomic operations on useswcnt ensure thread safety and
tp->lock guards access to tp->usesw and tp->counted, the broader lock
down_write(&block->cb_lock) is no longer required in tc_new_tfilter(),
and this resolves a performance regression caused by the filter counting
mechanism during parallel filter insertions.
The improvement can be demonstrated using the following script:
# cat insert_tc_rules.sh
tc qdisc add dev ens1f0np0 ingress
for i in $(seq 16); do
taskset -c $i tc -b rules_$i.txt &
done
wait
Each of rules_$i.txt files above includes 100000 tc filter rules to a
mlx5 driver NIC ens1f0np0.
Without this patch:
# time sh insert_tc_rules.sh
real 0m50.780s
user 0m23.556s
sys 4m13.032s
With this patch:
# time sh insert_tc_rules.sh
real 0m17.718s
user 0m7.807s
sys 3m45.050s
Fixes: 047f340b36 ("net: sched: make skip_sw actually skip software")
Reported-by: Shuang Li <shuali@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Reviewed-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 7a4b614063 ]
The conntrack entry is already public, there is a small chance that another
CPU is handling a packet in reply direction and racing with the tcp state
update.
Move this under ct spinlock.
This is done once, when ct is about to be offloaded, so this should
not result in a noticeable performance hit.
Fixes: 8437a6209f ("netfilter: nft_flow_offload: set liberal tracking mode for tcp")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 8d738c1869 ]
The existing rbtree implementation uses singleton elements to represent
ranges, however, userspace provides a set size according to the number
of ranges in the set.
Adjust provided userspace set size to the number of singleton elements
in the kernel by multiplying the range by two.
Check if the no-match all-zero element is already in the set, in such
case release one slot in the set size.
Fixes: 0ed6389c48 ("netfilter: nf_tables: rename set implementations")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit bc50835e83 ]
Lion Ackermann was able to create a UAF which can be abused for privilege
escalation with the following script
Step 1. create root qdisc
tc qdisc add dev lo root handle 1:0 drr
step2. a class for packet aggregation do demonstrate uaf
tc class add dev lo classid 1:1 drr
step3. a class for nesting
tc class add dev lo classid 1:2 drr
step4. a class to graft qdisc to
tc class add dev lo classid 1:3 drr
step5.
tc qdisc add dev lo parent 1:1 handle 2:0 plug limit 1024
step6.
tc qdisc add dev lo parent 1:2 handle 3:0 drr
step7.
tc class add dev lo classid 3:1 drr
step 8.
tc qdisc add dev lo parent 3:1 handle 4:0 pfifo
step 9. Display the class/qdisc layout
tc class ls dev lo
class drr 1:1 root leaf 2: quantum 64Kb
class drr 1:2 root leaf 3: quantum 64Kb
class drr 3:1 root leaf 4: quantum 64Kb
tc qdisc ls
qdisc drr 1: dev lo root refcnt 2
qdisc plug 2: dev lo parent 1:1
qdisc pfifo 4: dev lo parent 3:1 limit 1000p
qdisc drr 3: dev lo parent 1:2
step10. trigger the bug <=== prevented by this patch
tc qdisc replace dev lo parent 1:3 handle 4:0
step 11. Redisplay again the qdiscs/classes
tc class ls dev lo
class drr 1:1 root leaf 2: quantum 64Kb
class drr 1:2 root leaf 3: quantum 64Kb
class drr 1:3 root leaf 4: quantum 64Kb
class drr 3:1 root leaf 4: quantum 64Kb
tc qdisc ls
qdisc drr 1: dev lo root refcnt 2
qdisc plug 2: dev lo parent 1:1
qdisc pfifo 4: dev lo parent 3:1 refcnt 2 limit 1000p
qdisc drr 3: dev lo parent 1:2
Observe that a) parent for 4:0 does not change despite the replace request.
There can only be one parent. b) refcount has gone up by two for 4:0 and
c) both class 1:3 and 3:1 are pointing to it.
Step 12. send one packet to plug
echo "" | socat -u STDIN UDP4-DATAGRAM:127.0.0.1:8888,priority=$((0x10001))
step13. send one packet to the grafted fifo
echo "" | socat -u STDIN UDP4-DATAGRAM:127.0.0.1:8888,priority=$((0x10003))
step14. lets trigger the uaf
tc class delete dev lo classid 1:3
tc class delete dev lo classid 1:1
The semantics of "replace" is for a del/add _on the same node_ and not
a delete from one node(3:1) and add to another node (1:3) as in step10.
While we could "fix" with a more complex approach there could be
consequences to expectations so the patch takes the preventive approach of
"disallow such config".
Joint work with Lion Ackermann <nnamrec@gmail.com>
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250116013713.900000-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 12e070eb69 ]
The following trace can be seen if a device is being unregistered while
its number of channels are being modified.
DEBUG_LOCKS_WARN_ON(lock->magic != lock)
WARNING: CPU: 3 PID: 3754 at kernel/locking/mutex.c:564 __mutex_lock+0xc8a/0x1120
CPU: 3 UID: 0 PID: 3754 Comm: ethtool Not tainted 6.13.0-rc6+ #771
RIP: 0010:__mutex_lock+0xc8a/0x1120
Call Trace:
<TASK>
ethtool_check_max_channel+0x1ea/0x880
ethnl_set_channels+0x3c3/0xb10
ethnl_default_set_doit+0x306/0x650
genl_family_rcv_msg_doit+0x1e3/0x2c0
genl_rcv_msg+0x432/0x6f0
netlink_rcv_skb+0x13d/0x3b0
genl_rcv+0x28/0x40
netlink_unicast+0x42e/0x720
netlink_sendmsg+0x765/0xc20
__sys_sendto+0x3ac/0x420
__x64_sys_sendto+0xe0/0x1c0
do_syscall_64+0x95/0x180
entry_SYSCALL_64_after_hwframe+0x76/0x7e
This is because unregister_netdevice_many_notify might run before the
rtnl lock section of ethnl operations, eg. set_channels in the above
example. In this example the rss lock would be destroyed by the device
unregistration path before being used again, but in general running
ethnl operations while dismantle has started is not a good idea.
Fix this by denying any operation on devices being unregistered. A check
was already there in ethnl_ops_begin, but not wide enough.
Note that the same issue cannot be seen on the ioctl version
(__dev_ethtool) because the device reference is retrieved from within
the rtnl lock section there. Once dismantle started, the net device is
unlisted and no reference will be found.
Fixes: dde91ccfa2 ("ethtool: do not perform operations on net devices being unregistered")
Signed-off-by: Antoine Tenart <atenart@kernel.org>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Edward Cree <ecree.xilinx@gmail.com>
Link: https://patch.msgid.link/20250116092159.50890-1-atenart@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit d1f9f79fa2 ]
The following problem was encountered during stability test:
(NULL net_device): NAPI poll function process_backlog+0x0/0x530 \
returned 1, exceeding its budget of 0.
------------[ cut here ]------------
list_add double add: new=ffff88905f746f48, prev=ffff88905f746f48, \
next=ffff88905f746e40.
WARNING: CPU: 18 PID: 5462 at lib/list_debug.c:35 \
__list_add_valid_or_report+0xf3/0x130
CPU: 18 UID: 0 PID: 5462 Comm: ping Kdump: loaded Not tainted 6.13.0-rc7+
RIP: 0010:__list_add_valid_or_report+0xf3/0x130
Call Trace:
? __warn+0xcd/0x250
? __list_add_valid_or_report+0xf3/0x130
enqueue_to_backlog+0x923/0x1070
netif_rx_internal+0x92/0x2b0
__netif_rx+0x15/0x170
loopback_xmit+0x2ef/0x450
dev_hard_start_xmit+0x103/0x490
__dev_queue_xmit+0xeac/0x1950
ip_finish_output2+0x6cc/0x1620
ip_output+0x161/0x270
ip_push_pending_frames+0x155/0x1a0
raw_sendmsg+0xe13/0x1550
__sys_sendto+0x3bf/0x4e0
__x64_sys_sendto+0xdc/0x1b0
do_syscall_64+0x5b/0x170
entry_SYSCALL_64_after_hwframe+0x76/0x7e
The reproduction command is as follows:
sysctl -w net.core.dev_weight=0
ping 127.0.0.1
This is because when the napi's weight is set to 0, process_backlog() may
return 0 and clear the NAPI_STATE_SCHED bit of napi->state, causing this
napi to be re-polled in net_rx_action() until __do_softirq() times out.
Since the NAPI_STATE_SCHED bit has been cleared, napi_schedule_rps() can
be retriggered in enqueue_to_backlog(), causing this issue.
Making the napi's weight always non-zero solves this problem.
Triggering this issue requires system-wide admin (setting is
not namespaced).
Fixes: e387660545 ("[NET]: Fix sysctl net.core.dev_weight")
Fixes: 3d48b53fb2 ("net: dev_weight: TX/RX orthogonality")
Signed-off-by: Liu Jian <liujian56@huawei.com>
Link: https://patch.msgid.link/20250116143053.4146855-1-liujian56@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e361560a79 ]
The cited commit forgot to add netdev_rename_lock in one of the
error paths in dev_change_name().
Let's hold netdev_rename_lock before restoring the old dev->name.
Fixes: 0840556e5a ("net: Protect dev->name by seqlock.")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250115095545.52709-2-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Paolo Abeni
Paul Fertser
Dan Carpenter
Fedor Pchelkin
David Howells
Cong Wang
Quang Le
Eric Dumazet
Jakub Kicinski
Yan Zhai
Mazin Al Haddad
Gal Pressman
Steffen Klassert
Everest K.C
Matthieu Baerts (NGI0)
Pablo Neira Ayuso
Chuck Lever
Toke Høiland-Jørgensen
Jon Maloy
Michal Luczaj
Edward Cree
Sebastian Sewior
Florian Westphal
Jianbo Liu
Daniel Xu
Nikita Zhandarovich
Mahdi Arghavani
Xin Long
Jamal Hadi Salim
Antoine Tenart
Liu Jian
Kuniyuki Iwashima