[ Upstream commit df207de9d9 ]
Matt Dowling reported a weird UDP memory usage issue.
Under normal operation, the UDP memory usage reported in /proc/net/sockstat
remains close to zero. However, it occasionally spiked to 524,288 pages
and never dropped. Moreover, the value doubled when the application was
terminated. Finally, it caused intermittent packet drops.
We can reproduce the issue with the script below [0]:
1. /proc/net/sockstat reports 0 pages
# cat /proc/net/sockstat | grep UDP:
UDP: inuse 1 mem 0
2. Run the script till the report reaches 524,288
# python3 test.py & sleep 5
# cat /proc/net/sockstat | grep UDP:
UDP: inuse 3 mem 524288 <-- (INT_MAX + 1) >> PAGE_SHIFT
3. Kill the socket and confirm the number never drops
# pkill python3 && sleep 5
# cat /proc/net/sockstat | grep UDP:
UDP: inuse 1 mem 524288
4. (necessary since v6.0) Trigger proto_memory_pcpu_drain()
# python3 test.py & sleep 1 && pkill python3
5. The number doubles
# cat /proc/net/sockstat | grep UDP:
UDP: inuse 1 mem 1048577
The application set INT_MAX to SO_RCVBUF, which triggered an integer
overflow in udp_rmem_release().
When a socket is close()d, udp_destruct_common() purges its receive
queue and sums up skb->truesize in the queue. This total is calculated
and stored in a local unsigned integer variable.
The total size is then passed to udp_rmem_release() to adjust memory
accounting. However, because the function takes a signed integer
argument, the total size can wrap around, causing an overflow.
Then, the released amount is calculated as follows:
1) Add size to sk->sk_forward_alloc.
2) Round down sk->sk_forward_alloc to the nearest lower multiple of
PAGE_SIZE and assign it to amount.
3) Subtract amount from sk->sk_forward_alloc.
4) Pass amount >> PAGE_SHIFT to __sk_mem_reduce_allocated().
When the issue occurred, the total in udp_destruct_common() was 2147484480
(INT_MAX + 833), which was cast to -2147482816 in udp_rmem_release().
At 1) sk->sk_forward_alloc is changed from 3264 to -2147479552, and
2) sets -2147479552 to amount. 3) reverts the wraparound, so we don't
see a warning in inet_sock_destruct(). However, udp_memory_allocated
ends up doubling at 4).
Since commit 3cd3399dd7 ("net: implement per-cpu reserves for
memory_allocated"), memory usage no longer doubles immediately after
a socket is close()d because __sk_mem_reduce_allocated() caches the
amount in udp_memory_per_cpu_fw_alloc. However, the next time a UDP
socket receives a packet, the subtraction takes effect, causing UDP
memory usage to double.
This issue makes further memory allocation fail once the socket's
sk->sk_rmem_alloc exceeds net.ipv4.udp_rmem_min, resulting in packet
drops.
To prevent this issue, let's use unsigned int for the calculation and
call sk_forward_alloc_add() only once for the small delta.
Note that first_packet_length() also potentially has the same problem.
[0]:
from socket import *
SO_RCVBUFFORCE = 33
INT_MAX = (2 ** 31) - 1
s = socket(AF_INET, SOCK_DGRAM)
s.bind(('', 0))
s.setsockopt(SOL_SOCKET, SO_RCVBUFFORCE, INT_MAX)
c = socket(AF_INET, SOCK_DGRAM)
c.connect(s.getsockname())
data = b'a' * 100
while True:
c.send(data)
Fixes: f970bd9e3a ("udp: implement memory accounting helpers")
Reported-by: Matt Dowling <madowlin@amazon.com>
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20250401184501.67377-3-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 5a465a0da1 ]
__udp_enqueue_schedule_skb() has the following condition:
if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf)
goto drop;
sk->sk_rcvbuf is initialised by net.core.rmem_default and later can
be configured by SO_RCVBUF, which is limited by net.core.rmem_max,
or SO_RCVBUFFORCE.
If we set INT_MAX to sk->sk_rcvbuf, the condition is always false
as sk->sk_rmem_alloc is also signed int.
Then, the size of the incoming skb is added to sk->sk_rmem_alloc
unconditionally.
This results in integer overflow (possibly multiple times) on
sk->sk_rmem_alloc and allows a single socket to have skb up to
net.core.udp_mem[1].
For example, if we set a large value to udp_mem[1] and INT_MAX to
sk->sk_rcvbuf and flood packets to the socket, we can see multiple
overflows:
# cat /proc/net/sockstat | grep UDP:
UDP: inuse 3 mem 7956736 <-- (7956736 << 12) bytes > INT_MAX * 15
^- PAGE_SHIFT
# ss -uam
State Recv-Q ...
UNCONN -1757018048 ... <-- flipping the sign repeatedly
skmem:(r2537949248,rb2147483646,t0,tb212992,f1984,w0,o0,bl0,d0)
Previously, we had a boundary check for INT_MAX, which was removed by
commit 6a1f12dd85 ("udp: relax atomic operation on sk->sk_rmem_alloc").
A complete fix would be to revert it and cap the right operand by
INT_MAX:
rmem = atomic_add_return(size, &sk->sk_rmem_alloc);
if (rmem > min(size + (unsigned int)sk->sk_rcvbuf, INT_MAX))
goto uncharge_drop;
but we do not want to add the expensive atomic_add_return() back just
for the corner case.
Casting rmem to unsigned int prevents multiple wraparounds, but we still
allow a single wraparound.
# cat /proc/net/sockstat | grep UDP:
UDP: inuse 3 mem 524288 <-- (INT_MAX + 1) >> 12
# ss -uam
State Recv-Q ...
UNCONN -2147482816 ... <-- INT_MAX + 831 bytes
skmem:(r2147484480,rb2147483646,t0,tb212992,f3264,w0,o0,bl0,d14468947)
So, let's define rmem and rcvbuf as unsigned int and check skb->truesize
only when rcvbuf is large enough to lower the overflow possibility.
Note that we still have a small chance to see overflow if multiple skbs
to the same socket are processed on different core at the same time and
each size does not exceed the limit but the total size does.
Note also that we must ignore skb->truesize for a small buffer as
explained in commit 363dc73aca ("udp: be less conservative with
sock rmem accounting").
Fixes: 6a1f12dd85 ("udp: relax atomic operation on sk->sk_rmem_alloc")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20250401184501.67377-2-kuniyu@amazon.com
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 a502ea6fa9 ]
If a UDP socket changes its local address while it's receiving
datagrams, as a result of connect(), there is a period during which
a lookup operation might fail to find it, after the address is changed
but before the secondary hash (port and address) and the four-tuple
hash (local and remote ports and addresses) are updated.
Secondary hash chains were introduced by commit 30fff9231f ("udp:
bind() optimisation") and, as a result, a rehash operation became
needed to make a bound socket reachable again after a connect().
This operation was introduced by commit 719f835853 ("udp: add
rehash on connect()") which isn't however a complete fix: the
socket will be found once the rehashing completes, but not while
it's pending.
This is noticeable with a socat(1) server in UDP4-LISTEN mode, and a
client sending datagrams to it. After the server receives the first
datagram (cf. _xioopen_ipdgram_listen()), it issues a connect() to
the address of the sender, in order to set up a directed flow.
Now, if the client, running on a different CPU thread, happens to
send a (subsequent) datagram while the server's socket changes its
address, but is not rehashed yet, this will result in a failed
lookup and a port unreachable error delivered to the client, as
apparent from the following reproducer:
LEN=$(($(cat /proc/sys/net/core/wmem_default) / 4))
dd if=/dev/urandom bs=1 count=${LEN} of=tmp.in
while :; do
taskset -c 1 socat UDP4-LISTEN:1337,null-eof OPEN:tmp.out,create,trunc &
sleep 0.1 || sleep 1
taskset -c 2 socat OPEN:tmp.in UDP4:localhost:1337,shut-null
wait
done
where the client will eventually get ECONNREFUSED on a write()
(typically the second or third one of a given iteration):
2024/11/13 21:28:23 socat[46901] E write(6, 0x556db2e3c000, 8192): Connection refused
This issue was first observed as a seldom failure in Podman's tests
checking UDP functionality while using pasta(1) to connect the
container's network namespace, which leads us to a reproducer with
the lookup error resulting in an ICMP packet on a tap device:
LOCAL_ADDR="$(ip -j -4 addr show|jq -rM '.[] | .addr_info[0] | select(.scope == "global").local')"
while :; do
./pasta --config-net -p pasta.pcap -u 1337 socat UDP4-LISTEN:1337,null-eof OPEN:tmp.out,create,trunc &
sleep 0.2 || sleep 1
socat OPEN:tmp.in UDP4:${LOCAL_ADDR}:1337,shut-null
wait
cmp tmp.in tmp.out
done
Once this fails:
tmp.in tmp.out differ: char 8193, line 29
we can finally have a look at what's going on:
$ tshark -r pasta.pcap
1 0.000000 :: ? ff02::16 ICMPv6 110 Multicast Listener Report Message v2
2 0.168690 88.198.0.161 ? 88.198.0.164 UDP 8234 60260 ? 1337 Len=8192
3 0.168767 88.198.0.161 ? 88.198.0.164 UDP 8234 60260 ? 1337 Len=8192
4 0.168806 88.198.0.161 ? 88.198.0.164 UDP 8234 60260 ? 1337 Len=8192
5 0.168827 c6:47:05:8d:dc:04 ? Broadcast ARP 42 Who has 88.198.0.161? Tell 88.198.0.164
6 0.168851 9a:55:9a:55:9a:55 ? c6:47:05:8d:dc:04 ARP 42 88.198.0.161 is at 9a:55:9a:55:9a:55
7 0.168875 88.198.0.161 ? 88.198.0.164 UDP 8234 60260 ? 1337 Len=8192
8 0.168896 88.198.0.164 ? 88.198.0.161 ICMP 590 Destination unreachable (Port unreachable)
9 0.168926 88.198.0.161 ? 88.198.0.164 UDP 8234 60260 ? 1337 Len=8192
10 0.168959 88.198.0.161 ? 88.198.0.164 UDP 8234 60260 ? 1337 Len=8192
11 0.168989 88.198.0.161 ? 88.198.0.164 UDP 4138 60260 ? 1337 Len=4096
12 0.169010 88.198.0.161 ? 88.198.0.164 UDP 42 60260 ? 1337 Len=0
On the third datagram received, the network namespace of the container
initiates an ARP lookup to deliver the ICMP message.
In another variant of this reproducer, starting the client with:
strace -f pasta --config-net -u 1337 socat UDP4-LISTEN:1337,null-eof OPEN:tmp.out,create,trunc 2>strace.log &
and connecting to the socat server using a loopback address:
socat OPEN:tmp.in UDP4:localhost:1337,shut-null
we can more clearly observe a sendmmsg() call failing after the
first datagram is delivered:
[pid 278012] connect(173, 0x7fff96c95fc0, 16) = 0
[...]
[pid 278012] recvmmsg(173, 0x7fff96c96020, 1024, MSG_DONTWAIT, NULL) = -1 EAGAIN (Resource temporarily unavailable)
[pid 278012] sendmmsg(173, 0x561c5ad0a720, 1, MSG_NOSIGNAL) = 1
[...]
[pid 278012] sendmmsg(173, 0x561c5ad0a720, 1, MSG_NOSIGNAL) = -1 ECONNREFUSED (Connection refused)
and, somewhat confusingly, after a connect() on the same socket
succeeded.
Until commit 4cdeeee925 ("net: udp: prefer listeners bound to an
address"), the race between receive address change and lookup didn't
actually cause visible issues, because, once the lookup based on the
secondary hash chain failed, we would still attempt a lookup based on
the primary hash (destination port only), and find the socket with the
outdated secondary hash.
That change, however, dropped port-only lookups altogether, as side
effect, making the race visible.
To fix this, while avoiding the need to make address changes and
rehash atomic against lookups, reintroduce primary hash lookups as
fallback, if lookups based on four-tuple and secondary hashes fail.
To this end, introduce a simplified lookup implementation, which
doesn't take care of SO_REUSEPORT groups: if we have one, there are
multiple sockets that would match the four-tuple or secondary hash,
meaning that we can't run into this race at all.
v2:
- instead of synchronising lookup operations against address change
plus rehash, reintroduce a simplified version of the original
primary hash lookup as fallback
v1:
- fix build with CONFIG_IPV6=n: add ifdef around sk_v6_rcv_saddr
usage (Kuniyuki Iwashima)
- directly use sk_rcv_saddr for IPv4 receive addresses instead of
fetching inet_rcv_saddr (Kuniyuki Iwashima)
- move inet_update_saddr() to inet_hashtables.h and use that
to set IPv4/IPv6 addresses as suitable (Kuniyuki Iwashima)
- rebase onto net-next, update commit message accordingly
Reported-by: Ed Santiago <santiago@redhat.com>
Link: https://github.com/containers/podman/issues/24147
Analysed-by: David Gibson <david@gibson.dropbear.id.au>
Fixes: 30fff9231f ("udp: bind() optimisation")
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.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>
If:
1) the user requested USO, but
2) there is not enough payload for GSO to kick in, and
3) the egress device doesn't offer checksum offload, then
we want to compute the L4 checksum in software early on.
In the case when we are not taking the GSO path, but it has been requested,
the software checksum fallback in skb_segment doesn't get a chance to
compute the full checksum, if the egress device can't do it. As a result we
end up sending UDP datagrams with only a partial checksum filled in, which
the peer will discard.
Fixes: 10154dbded ("udp: Allow GSO transmit from devices with no checksum offload")
Reported-by: Ivan Babrou <ivan@cloudflare.com>
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20241011-uso-swcsum-fixup-v2-1-6e1ddc199af9@cloudflare.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Unmask the upper DSCP bits when performing source validation for
multicast packets during early demux. In the future, this will allow us
to perform the FIB lookup which is performed as part of source
validation according to the full DSCP value.
No functional changes intended since the upper DSCP bits are masked when
comparing against the TOS selectors in FIB rules and routes.
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Guillaume Nault <gnault@redhat.com>
Acked-by: Florian Westphal <fw@strlen.de>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://patch.msgid.link/20240821125251.1571445-12-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Cross-merge networking fixes after downstream PR.
Conflicts:
net/sched/act_ct.c
26488172b0 ("net/sched: Fix UAF when resolving a clash")
3abbd7ed8b ("act_ct: prepare for stolen verdict coming from conntrack and nat engine")
No adjacent changes.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Today sending a UDP GSO packet from a TUN device results in an EIO error:
import fcntl, os, struct
from socket import *
TUNSETIFF = 0x400454CA
IFF_TUN = 0x0001
IFF_NO_PI = 0x1000
UDP_SEGMENT = 103
tun_fd = os.open("/dev/net/tun", os.O_RDWR)
ifr = struct.pack("16sH", b"tun0", IFF_TUN | IFF_NO_PI)
fcntl.ioctl(tun_fd, TUNSETIFF, ifr)
os.system("ip addr add 192.0.2.1/24 dev tun0")
os.system("ip link set dev tun0 up")
s = socket(AF_INET, SOCK_DGRAM)
s.setsockopt(SOL_UDP, UDP_SEGMENT, 1200)
s.sendto(b"x" * 3000, ("192.0.2.2", 9)) # EIO
This is due to a check in the udp stack if the egress device offers
checksum offload. While TUN/TAP devices, by default, don't advertise this
capability because it requires support from the TUN/TAP reader.
However, the GSO stack has a software fallback for checksum calculation,
which we can use. This way we don't force UDP_SEGMENT users to handle the
EIO error and implement a segmentation fallback.
Lift the restriction so that UDP_SEGMENT can be used with any egress
device. We also need to adjust the UDP GSO code to match the GSO stack
expectation about ip_summed field, as set in commit 8d63bee643 ("net:
avoid skb_warn_bad_offload false positives on UFO"). Otherwise we will hit
the bad offload check.
Users should, however, expect a potential performance impact when
batch-sending packets with UDP_SEGMENT without checksum offload on the
egress device. In such case the packet payload is read twice: first during
the sendmsg syscall when copying data from user memory, and then in the GSO
stack for checksum computation. This double memory read can be less
efficient than a regular sendmsg where the checksum is calculated during
the initial data copy from user memory.
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240626-linux-udpgso-v2-1-422dfcbd6b48@cloudflare.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
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>
Commits a602456 ("udp: Add GRO functions to UDP socket") and 57c67ff ("udp:
additional GRO support") introduce incorrect usage of {ip,ipv6}_hdr in the
complete phase of gro. The functions always return skb->network_header,
which in the case of encapsulated packets at the gro complete phase, is
always set to the innermost L3 of the packet. That means that calling
{ip,ipv6}_hdr for skbs which completed the GRO receive phase (both in
gro_list and *_gro_complete) when parsing an encapsulated packet's _outer_
L3/L4 may return an unexpected value.
This incorrect usage leads to a bug in GRO's UDP socket lookup.
udp{4,6}_lib_lookup_skb functions use ip_hdr/ipv6_hdr respectively. These
*_hdr functions return network_header which will point to the innermost L3,
resulting in the wrong offset being used in __udp{4,6}_lib_lookup with
encapsulated packets.
This patch adds network_offset and inner_network_offset to napi_gro_cb, and
makes sure both are set correctly.
To fix the issue, network_offsets union is used inside napi_gro_cb, in
which both the outer and the inner network offsets are saved.
Reproduction example:
Endpoint configuration example (fou + local address bind)
# ip fou add port 6666 ipproto 4
# ip link add name tun1 type ipip remote 2.2.2.1 local 2.2.2.2 encap fou encap-dport 5555 encap-sport 6666 mode ipip
# ip link set tun1 up
# ip a add 1.1.1.2/24 dev tun1
Netperf TCP_STREAM result on net-next before patch is applied:
net-next main, GRO enabled:
$ netperf -H 1.1.1.2 -t TCP_STREAM -l 5
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
131072 16384 16384 5.28 2.37
net-next main, GRO disabled:
$ netperf -H 1.1.1.2 -t TCP_STREAM -l 5
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
131072 16384 16384 5.01 2745.06
patch applied, GRO enabled:
$ netperf -H 1.1.1.2 -t TCP_STREAM -l 5
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
131072 16384 16384 5.01 2877.38
Fixes: a6024562ff ("udp: Add GRO functions to UDP socket")
Signed-off-by: Richard Gobert <richardbgobert@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The UDP_ENCAP_ESPINUDP_NON_IKE mode, introduced into the Linux kernel
in 2004 [2], has remained inactive and obsolete for an extended period.
This mode was originally defined in an early version of an IETF draft
[1] from 2001. By the time it was integrated into the kernel in 2004 [2],
it had already been replaced by UDP_ENCAP_ESPINUDP [3] in later
versions of draft-ietf-ipsec-udp-encaps, particularly in version 06.
Over time, UDP_ENCAP_ESPINUDP_NON_IKE has lost its relevance, with no
known use cases.
With this commit, we remove support for UDP_ENCAP_ESPINUDP_NON_IKE,
simplifying the codebase and eliminating unnecessary complexity.
Kernel will return an error -ENOPROTOOPT if the userspace tries to set
this option.
References:
[1] https://datatracker.ietf.org/doc/html/draft-ietf-ipsec-udp-encaps-00.txt
[2] Commit that added UDP_ENCAP_ESPINUDP_NON_IKE to the Linux historic
repository.
Author: Andreas Gruenbacher <agruen@suse.de>
Date: Fri Apr 9 01:47:47 2004 -0700
[IPSEC]: Support draft-ietf-ipsec-udp-encaps-00/01, some ipec impls need it.
[3] Commit that added UDP_ENCAP_ESPINUDP to the Linux historic
repository.
Author: Derek Atkins <derek@ihtfp.com>
Date: Wed Apr 2 13:21:02 2003 -0800
[IPSEC]: Implement UDP Encapsulation framework.
Signed-off-by: Antony Antony <antony.antony@secunet.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
We've observed a 7-12% performance regression in iperf3 UDP ipv4 and
ipv6 tests with multiple sockets on Zen3 cpus, which we traced back to
commit f0ea27e7bf ("udp: re-score reuseport groups when connected
sockets are present"). The failing tests were those that would spawn
UDP sockets per-cpu on systems that have a high number of cpus.
Unsurprisingly, it is not caused by the extra re-scoring of the reused
socket, but due to the compiler no longer inlining compute_score, once
it has the extra call site in udp4_lib_lookup2. This is augmented by
the "Safe RET" mitigation for SRSO, needed in our Zen3 cpus.
We could just explicitly inline it, but compute_score() is quite a large
function, around 300b. Inlining in two sites would almost double
udp4_lib_lookup2, which is a silly thing to do just to workaround a
mitigation. Instead, this patch shuffles the code a bit to avoid the
multiple calls to compute_score. Since it is a static function used in
one spot, the compiler can safely fold it in, as it did before, without
increasing the text size.
With this patch applied I ran my original iperf3 testcases. The failing
cases all looked like this (ipv4):
iperf3 -c 127.0.0.1 --udp -4 -f K -b $R -l 8920 -t 30 -i 5 -P 64 -O 2
where $R is either 1G/10G/0 (max, unlimited). I ran 3 times each.
baseline is v6.9-rc3. harmean == harmonic mean; CV == coefficient of
variation.
ipv4:
1G 10G MAX
HARMEAN (CV) HARMEAN (CV) HARMEAN (CV)
baseline 1743852.66(0.0208) 1725933.02(0.0167) 1705203.78(0.0386)
patched 1968727.61(0.0035) 1962283.22(0.0195) 1923853.50(0.0256)
ipv6:
1G 10G MAX
HARMEAN (CV) HARMEAN (CV) HARMEAN (CV)
baseline 1729020.03(0.0028) 1691704.49(0.0243) 1692251.34(0.0083)
patched 1900422.19(0.0067) 1900968.01(0.0067) 1568532.72(0.1519)
This restores the performance we had before the change above with this
benchmark. We obviously don't expect any real impact when mitigations
are disabled, but just to be sure it also doesn't regresses:
mitigations=off ipv4:
1G 10G MAX
HARMEAN (CV) HARMEAN (CV) HARMEAN (CV)
baseline 3230279.97(0.0066) 3229320.91(0.0060) 2605693.19(0.0697)
patched 3242802.36(0.0073) 3239310.71(0.0035) 2502427.19(0.0882)
Cc: Lorenz Bauer <lmb@isovalent.com>
Fixes: f0ea27e7bf ("udp: re-score reuseport groups when connected sockets are present")
Signed-off-by: Gabriel Krisman Bertazi <krisman@suse.de>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
While looking at UDP receive performance, I saw sk_wake_async()
was no longer inlined.
This matters at least on AMD Zen1-4 platforms (see SRSO)
This might be because rcu_read_lock() and rcu_read_unlock()
are no longer nops in recent kernels ?
Add sk_wake_async_rcu() variant, which must be called from
contexts already holding rcu lock.
As SOCK_FASYNC is deprecated in modern days, use unlikely()
to give a hint to the compiler.
sk_wake_async_rcu() is properly inlined from
__udp_enqueue_schedule_skb() and sock_def_readable().
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240328144032.1864988-5-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
sock_def_readable() is quite expensive (particularly
when ep_poll_callback() is in the picture).
We must call sk->sk_data_ready() when :
- receive queue was empty, or
- SO_PEEK_OFF is enabled on the socket, or
- sk->sk_data_ready is not sock_def_readable.
We still need to call sk_wake_async().
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Link: https://lore.kernel.org/r/20240328144032.1864988-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When rx-udp-gro-forwarding is enabled UDP packets might be GROed when
being forwarded. If such packets might land in a tunnel this can cause
various issues and udp_gro_receive makes sure this isn't the case by
looking for a matching socket. This is performed in
udp4/6_gro_lookup_skb but only in the current netns. This is an issue
with tunneled packets when the endpoint is in another netns. In such
cases the packets will be GROed at the UDP level, which leads to various
issues later on. The same thing can happen with rx-gro-list.
We saw this with geneve packets being GROed at the UDP level. In such
case gso_size is set; later the packet goes through the geneve rx path,
the geneve header is pulled, the offset are adjusted and frag_list skbs
are not adjusted with regard to geneve. When those skbs hit
skb_fragment, it will misbehave. Different outcomes are possible
depending on what the GROed skbs look like; from corrupted packets to
kernel crashes.
One example is a BUG_ON[1] triggered in skb_segment while processing the
frag_list. Because gso_size is wrong (geneve header was pulled)
skb_segment thinks there is "geneve header size" of data in frag_list,
although it's in fact the next packet. The BUG_ON itself has nothing to
do with the issue. This is only one of the potential issues.
Looking up for a matching socket in udp_gro_receive is fragile: the
lookup could be extended to all netns (not speaking about performances)
but nothing prevents those packets from being modified in between and we
could still not find a matching socket. It's OK to keep the current
logic there as it should cover most cases but we also need to make sure
we handle tunnel packets being GROed too early.
This is done by extending the checks in udp_unexpected_gso: GSO packets
lacking the SKB_GSO_UDP_TUNNEL/_CSUM bits and landing in a tunnel must
be segmented.
[1] kernel BUG at net/core/skbuff.c:4408!
RIP: 0010:skb_segment+0xd2a/0xf70
__udp_gso_segment+0xaa/0x560
Fixes: 9fd1ff5d2a ("udp: Support UDP fraglist GRO/GSO.")
Fixes: 36707061d6 ("udp: allow forwarding of plain (non-fraglisted) UDP GRO packets")
Signed-off-by: Antoine Tenart <atenart@kernel.org>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The 'len' variable can't be negative when assigned the result of
'min_t' because all 'min_t' parameters are cast to unsigned int,
and then the minimum one is chosen.
To fix the logic, check 'len' as read from 'optlen',
where the types of relevant variables are (signed) int.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Gavrilov Ilia <Ilia.Gavrilov@infotecs.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
We want to re-organize the struct sock layout. The sk_peek_off
field location is problematic, as most protocols want it in the
RX read area, while UDP wants it on a cacheline different from
sk_receive_queue.
Create a local (inside udp_sock) copy of the 'peek offset is enabled'
flag and place it inside the same cacheline of reader_queue.
Check such flag before reading sk_peek_off. This will save potential
false sharing and cache misses in the fast-path.
Tested under UDP flood with small packets. The struct sock layout
update causes a 4% performance drop, and this patch restores completely
the original tput.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/67ab679c15fbf49fa05b3ffe05d91c47ab84f147.1708426665.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
There is a bug in the bpf_iter_udp_batch() function that stops
the userspace from making forward progress.
The case that triggers the bug is the userspace passed in
a very small read buffer. When the bpf prog does bpf_seq_printf,
the userspace read buffer is not enough to capture the whole bucket.
When the read buffer is not large enough, the kernel will remember
the offset of the bucket in iter->offset such that the next userspace
read() can continue from where it left off.
The kernel will skip the number (== "iter->offset") of sockets in
the next read(). However, the code directly decrements the
"--iter->offset". This is incorrect because the next read() may
not consume the whole bucket either and then the next-next read()
will start from offset 0. The net effect is the userspace will
keep reading from the beginning of a bucket and the process will
never finish. "iter->offset" must always go forward until the
whole bucket is consumed.
This patch fixes it by using a local variable "resume_offset"
and "resume_bucket". "iter->offset" is always reset to 0 before
it may be used. "iter->offset" will be advanced to the
"resume_offset" when it continues from the "resume_bucket" (i.e.
"state->bucket == resume_bucket"). This brings it closer to
the bpf_iter_tcp's offset handling which does not suffer
the same bug.
Cc: Aditi Ghag <aditi.ghag@isovalent.com>
Fixes: c96dac8d36 ("bpf: udp: Implement batching for sockets iterator")
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Reviewed-by: Aditi Ghag <aditi.ghag@isovalent.com>
Link: https://lore.kernel.org/r/20240112190530.3751661-3-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The current logic is to use a default size 16 to batch the whole bucket.
If it is too small, it will retry with a larger batch size.
The current code accidentally does a state->bucket-- before retrying.
This goes back to retry with the previous bucket which has already
been done. This patch fixed it.
It is hard to create a selftest. I added a WARN_ON(state->bucket < 0),
forced a particular port to be hashed to the first bucket,
created >16 sockets, and observed the for-loop went back
to the "-1" bucket.
Cc: Aditi Ghag <aditi.ghag@isovalent.com>
Fixes: c96dac8d36 ("bpf: udp: Implement batching for sockets iterator")
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Reviewed-by: Aditi Ghag <aditi.ghag@isovalent.com>
Link: https://lore.kernel.org/r/20240112190530.3751661-2-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Steffen Klassert says:
====================
pull request (net-next): ipsec-next 2023-10-28
1) Remove unused function declarations of xfrm4_extract_input and
xfrm6_extract_input. From Yue Haibing.
2) Annotate struct xfrm_sec_ctx with __counted_by.
From Kees Cook.
3) Support GRO decapsulation for ESP in UDP encapsulation.
From Antony Antony et all.
4) Replace the xfrm session decode with flow dissector.
From Florian Westphal.
5) Fix a use after free in __xfrm6_udp_encap_rcv.
6) Fix the layer 4 flowi decoding.
From Florian Westphal.
* tag 'ipsec-next-2023-10-28' of git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec-next:
xfrm: policy: fix layer 4 flowi decoding
xfrm Fix use after free in __xfrm6_udp_encap_rcv.
xfrm: policy: replace session decode with flow dissector
xfrm: move mark and oif flowi decode into common code
xfrm: pass struct net to xfrm_decode_session wrappers
xfrm: Support GRO for IPv6 ESP in UDP encapsulation
xfrm: Support GRO for IPv4 ESP in UDP encapsulation
xfrm: Use the XFRM_GRO to indicate a GRO call on input
xfrm: Annotate struct xfrm_sec_ctx with __counted_by
xfrm: Remove unused function declarations
====================
Link: https://lore.kernel.org/r/20231028084328.3119236-1-steffen.klassert@secunet.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Daniel Borkmann says:
====================
pull-request: bpf-next 2023-10-16
We've added 90 non-merge commits during the last 25 day(s) which contain
a total of 120 files changed, 3519 insertions(+), 895 deletions(-).
The main changes are:
1) Add missed stats for kprobes to retrieve the number of missed kprobe
executions and subsequent executions of BPF programs, from Jiri Olsa.
2) Add cgroup BPF sockaddr hooks for unix sockets. The use case is
for systemd to reimplement the LogNamespace feature which allows
running multiple instances of systemd-journald to process the logs
of different services, from Daan De Meyer.
3) Implement BPF CPUv4 support for s390x BPF JIT, from Ilya Leoshkevich.
4) Improve BPF verifier log output for scalar registers to better
disambiguate their internal state wrt defaults vs min/max values
matching, from Andrii Nakryiko.
5) Extend the BPF fib lookup helpers for IPv4/IPv6 to support retrieving
the source IP address with a new BPF_FIB_LOOKUP_SRC flag,
from Martynas Pumputis.
6) Add support for open-coded task_vma iterator to help with symbolization
for BPF-collected user stacks, from Dave Marchevsky.
7) Add libbpf getters for accessing individual BPF ring buffers which
is useful for polling them individually, for example, from Martin Kelly.
8) Extend AF_XDP selftests to validate the SHARED_UMEM feature,
from Tushar Vyavahare.
9) Improve BPF selftests cross-building support for riscv arch,
from Björn Töpel.
10) Add the ability to pin a BPF timer to the same calling CPU,
from David Vernet.
11) Fix libbpf's bpf_tracing.h macros for riscv to use the generic
implementation of PT_REGS_SYSCALL_REGS() to access syscall arguments,
from Alexandre Ghiti.
12) Extend libbpf to support symbol versioning for uprobes, from Hengqi Chen.
13) Fix bpftool's skeleton code generation to guarantee that ELF data
is 8 byte aligned, from Ian Rogers.
14) Inherit system-wide cpu_mitigations_off() setting for Spectre v1/v4
security mitigations in BPF verifier, from Yafang Shao.
15) Annotate struct bpf_stack_map with __counted_by attribute to prepare
BPF side for upcoming __counted_by compiler support, from Kees Cook.
* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: (90 commits)
bpf: Ensure proper register state printing for cond jumps
bpf: Disambiguate SCALAR register state output in verifier logs
selftests/bpf: Make align selftests more robust
selftests/bpf: Improve missed_kprobe_recursion test robustness
selftests/bpf: Improve percpu_alloc test robustness
selftests/bpf: Add tests for open-coded task_vma iter
bpf: Introduce task_vma open-coded iterator kfuncs
selftests/bpf: Rename bpf_iter_task_vma.c to bpf_iter_task_vmas.c
bpf: Don't explicitly emit BTF for struct btf_iter_num
bpf: Change syscall_nr type to int in struct syscall_tp_t
net/bpf: Avoid unused "sin_addr_len" warning when CONFIG_CGROUP_BPF is not set
bpf: Avoid unnecessary audit log for CPU security mitigations
selftests/bpf: Add tests for cgroup unix socket address hooks
selftests/bpf: Make sure mount directory exists
documentation/bpf: Document cgroup unix socket address hooks
bpftool: Add support for cgroup unix socket address hooks
libbpf: Add support for cgroup unix socket address hooks
bpf: Implement cgroup sockaddr hooks for unix sockets
bpf: Add bpf_sock_addr_set_sun_path() to allow writing unix sockaddr from bpf
bpf: Propagate modified uaddrlen from cgroup sockaddr programs
...
====================
Link: https://lore.kernel.org/r/20231016204803.30153-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
As prep for adding unix socket support to the cgroup sockaddr hooks,
let's propagate the sockaddr length back to the caller after running
a bpf cgroup sockaddr hook program. While not important for AF_INET or
AF_INET6, the sockaddr length is important when working with AF_UNIX
sockaddrs as the size of the sockaddr cannot be determined just from the
address family or the sockaddr's contents.
__cgroup_bpf_run_filter_sock_addr() is modified to take the uaddrlen as
an input/output argument. After running the program, the modified sockaddr
length is stored in the uaddrlen pointer.
Signed-off-by: Daan De Meyer <daan.j.demeyer@gmail.com>
Link: https://lore.kernel.org/r/20231011185113.140426-3-daan.j.demeyer@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
This patch enables the GRO codepath for IPv6 ESP in UDP encapsulated
packets. Decapsulation happens at L2 and saves a full round through
the stack for each packet. This is also needed to support HW offload
for ESP in UDP encapsulation.
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Co-developed-by: Antony Antony <antony.antony@secunet.com>
Signed-off-by: Antony Antony <antony.antony@secunet.com>
Reviewed-by: Eyal Birger <eyal.birger@gmail.com>
This patch enables the GRO codepath for IPv4 ESP in UDP encapsulated
packets. Decapsulation happens at L2 and saves a full round through
the stack for each packet. This is also needed to support HW offload
for ESP in UDP encapsulation.
Enabling this would imporove performance for ESP in UDP datapath, i.e
IPsec with NAT in between.
By default GRP for ESP-in-UDP is disabled for UDP sockets.
To enable this feature for an ESP socket, the following two options
need to be set:
1. enable ESP-in-UDP: (this is already set by an IKE daemon).
int type = UDP_ENCAP_ESPINUDP;
setsockopt(fd, SOL_UDP, UDP_ENCAP, &type, sizeof(type));
2. To enable GRO for ESP in UDP socket:
type = true;
setsockopt(fd, SOL_UDP, UDP_GRO, &type, sizeof(type));
Enabling ESP-in-UDP has the side effect of preventing the Linux stack from
seeing ESP packets at the L3 (when ESP OFFLOAD is disabled), as packets are
immediately decapsulated from UDP and decrypted.
This change may affect nftable rules that match on ESP packets at L3.
Also tcpdump won't see the ESP packet.
Developers/admins are advised to review and adapt any nftable rules
accordingly before enabling this feature to prevent potential rule breakage.
Also tcpdump will not see from ESP packets from a ESP in UDP flow, when this
is enabled.
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Co-developed-by: Antony Antony <antony.antony@secunet.com>
Signed-off-by: Antony Antony <antony.antony@secunet.com>
Reviewed-by: Eyal Birger <eyal.birger@gmail.com>
Add missing annotations to inet->mc_index and inet->mc_addr
to fix data-races.
getsockopt(IP_MULTICAST_IF) can be lockless.
setsockopt() side is left for later.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add missing READ_ONCE() annotations when reading inet->uc_index
Implementing getsockopt(IP_UNICAST_IF) locklessly seems possible,
the setsockopt() part might not be possible at the moment.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
inet->pmtudisc can be read locklessly.
Implement proper lockless reads and writes to inet->pmtudisc
ip_sock_set_mtu_discover() can now be called from arbitrary
contexts.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
udp->pcflag, udp->pcslen and udp->pcrlen reads/writes are racy.
Move udp->pcflag to udp->udp_flags for atomicity,
and add READ_ONCE()/WRITE_ONCE() annotations for pcslen and pcrlen.
Fixes: ba4e58eca8 ("[NET]: Supporting UDP-Lite (RFC 3828) in Linux")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Move udp->encap_enabled to udp->udp_flags.
Add udp_test_and_set_bit() helper to allow lockless
udp_tunnel_encap_enable() implementation.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
These are read locklessly, move them to udp_flags to fix data-races.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
