[ Upstream commit eb44ad4e63 ]
This field can be read or written without socket lock being held.
Add annotations to avoid load-store tearing.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 73ed8e0338 ]
cookie_init_timestamp() is supposed to return a 64bit timestamp
suitable for both TSval determination and setting of skb->tstamp.
Unfortunately it uses 32bit fields and overflows after
2^32 * 10^6 nsec (~49 days) of uptime.
Generated TSval are still correct, but skb->tstamp might be set
far away in the past, potentially confusing other layers.
tcp_ns_to_ts() is changed to return a full 64bit value,
ts and ts_now variables are changed to u64 type,
and TSMASK is removed in favor of shifts operations.
While we are at it, change this sequence:
ts >>= TSBITS;
ts--;
ts <<= TSBITS;
ts |= options;
to:
ts -= (1UL << TSBITS);
Fixes: 9a568de481 ("tcp: switch TCP TS option (RFC 7323) to 1ms clock")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit a135798e6e ]
tcp_init_metrics() only wants to get metrics if they were
previously stored in the cache. Creating an entry is adding
useless costs, especially when tcp_no_metrics_save is set.
Fixes: 51c5d0c4b1 ("tcp: Maintain dynamic metrics in local cache.")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 081480014a ]
We need to set tp->snd_ssthresh to TCP_INFINITE_SSTHRESH
in the case tcp_get_metrics() fails for some reason.
Fixes: 9ad7c049f0 ("tcp: RFC2988bis + taking RTT sample from 3WHS for the passive open side")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit cbc3a15322 ]
When removing an item from RCU protected list, we must prevent
store-tearing, using rcu_assign_pointer() or WRITE_ONCE().
Fixes: 04f721c671 ("tcp_metrics: Rewrite tcp_metrics_flush_all")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e326578a21 ]
For passive TCP Fast Open sockets that had SYN/ACK timeout and did not
send more data in SYN_RECV, upon receiving the final ACK in 3WHS, the
congestion state may awkwardly stay in CA_Loss mode unless the CA state
was undone due to TCP timestamp checks. However, if
tcp_rcv_synrecv_state_fastopen() decides not to undo, then we should
enter CA_Open, because at that point we have received an ACK covering
the retransmitted SYNACKs. Currently, the icsk_ca_state is only set to
CA_Open after we receive an ACK for a data-packet. This is because
tcp_ack does not call tcp_fastretrans_alert (and tcp_process_loss) if
!prior_packets
Note that tcp_process_loss() calls tcp_try_undo_recovery(), so having
tcp_rcv_synrecv_state_fastopen() decide that if we're in CA_Loss we
should call tcp_try_undo_recovery() is consistent with that, and
low risk.
Fixes: dad8cea7ad ("tcp: fix TFO SYNACK undo to avoid double-timestamp-undo")
Signed-off-by: Aananth V <aananthv@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 6d5a12eb91 ]
UDP_ENCAP_ESPINUDP_NON_IKE setsockopt() writes over up->encap_rcv
while other cpus read it.
Fixes: 067b207b28 ("[UDP]: Cleanup UDP encapsulation code")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit d2a0fc372a ]
This commit fix wrong RTO timeout when received SACK reneging.
When an ACK arrived pointing to a SACK reneging, tcp_check_sack_reneging()
will rearm the RTO timer for min(1/2*srtt, 10ms) into to the future.
But since the commit 62d9f1a694 ("tcp: fix TLP timer not set when
CA_STATE changes from DISORDER to OPEN") merged, the tcp_set_xmit_timer()
is moved after tcp_fastretrans_alert()(which do the SACK reneging check),
so the RTO timeout will be overwrited by tcp_set_xmit_timer() with
icsk_rto instead of 1/2*srtt.
Here is a packetdrill script to check this bug:
0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+0 bind(3, ..., ...) = 0
+0 listen(3, 1) = 0
// simulate srtt to 100ms
+0 < S 0:0(0) win 32792 <mss 1000, sackOK,nop,nop,nop,wscale 7>
+0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 7>
+.1 < . 1:1(0) ack 1 win 1024
+0 accept(3, ..., ...) = 4
+0 write(4, ..., 10000) = 10000
+0 > P. 1:10001(10000) ack 1
// inject sack
+.1 < . 1:1(0) ack 1 win 257 <sack 1001:10001,nop,nop>
+0 > . 1:1001(1000) ack 1
// inject sack reneging
+.1 < . 1:1(0) ack 1001 win 257 <sack 9001:10001,nop,nop>
// we expect rto fired in 1/2*srtt (50ms)
+.05 > . 1001:2001(1000) ack 1
This fix remove the FLAG_SET_XMIT_TIMER from ack_flag when
tcp_check_sack_reneging() set RTO timer with 1/2*srtt to avoid
being overwrited later.
Fixes: 62d9f1a694 ("tcp: fix TLP timer not set when CA_STATE changes from DISORDER to OPEN")
Signed-off-by: Fred Chen <fred.chenchen03@gmail.com>
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 68b54aeff8 ]
In the blamed commit below, I completely forgot to release the acquired
resources before erroring out in the TCP BPF code, as reported by Dan.
Address the issues by replacing the bogus return with a jump to the
relevant cleanup code.
Fixes: 419ce133ab ("tcp: allow again tcp_disconnect() when threads are waiting")
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Jakub Sitnicki <jakub@cloudflare.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/r/8f99194c698bcef12666f0a9a999c58f8b1cb52c.1697557782.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 419ce133ab ]
As reported by Tom, .NET and applications build on top of it rely
on connect(AF_UNSPEC) to async cancel pending I/O operations on TCP
socket.
The blamed commit below caused a regression, as such cancellation
can now fail.
As suggested by Eric, this change addresses the problem explicitly
causing blocking I/O operation to terminate immediately (with an error)
when a concurrent disconnect() is executed.
Instead of tracking the number of threads blocked on a given socket,
track the number of disconnect() issued on such socket. If such counter
changes after a blocking operation releasing and re-acquiring the socket
lock, error out the current operation.
Fixes: 4faeee0cf8 ("tcp: deny tcp_disconnect() when threads are waiting")
Reported-by: Tom Deseyn <tdeseyn@redhat.com>
Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1886305
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/f3b95e47e3dbed840960548aebaa8d954372db41.1697008693.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 4b2b606075 ]
After deleting an interface address in fib_del_ifaddr(), the function
scans the fib_info list for stray entries and calls fib_flush() and
fib_table_flush(). Then the stray entries will be deleted silently and no
RTM_DELROUTE notification will be sent.
This lack of notification can make routing daemons, or monitor like
`ip monitor route` miss the routing changes. e.g.
+ ip link add dummy1 type dummy
+ ip link add dummy2 type dummy
+ ip link set dummy1 up
+ ip link set dummy2 up
+ ip addr add 192.168.5.5/24 dev dummy1
+ ip route add 7.7.7.0/24 dev dummy2 src 192.168.5.5
+ ip -4 route
7.7.7.0/24 dev dummy2 scope link src 192.168.5.5
192.168.5.0/24 dev dummy1 proto kernel scope link src 192.168.5.5
+ ip monitor route
+ ip addr del 192.168.5.5/24 dev dummy1
Deleted 192.168.5.0/24 dev dummy1 proto kernel scope link src 192.168.5.5
Deleted broadcast 192.168.5.255 dev dummy1 table local proto kernel scope link src 192.168.5.5
Deleted local 192.168.5.5 dev dummy1 table local proto kernel scope host src 192.168.5.5
As Ido reminded, fib_table_flush() isn't only called when an address is
deleted, but also when an interface is deleted or put down. The lack of
notification in these cases is deliberate. And commit 7c6bb7d2fa
("net/ipv6: Add knob to skip DELROUTE message on device down") introduced
a sysctl to make IPv6 behave like IPv4 in this regard. So we can't send
the route delete notify blindly in fib_table_flush().
To fix this issue, let's add a new flag in "struct fib_info" to track the
deleted prefer source address routes, and only send notify for them.
After update:
+ ip monitor route
+ ip addr del 192.168.5.5/24 dev dummy1
Deleted 192.168.5.0/24 dev dummy1 proto kernel scope link src 192.168.5.5
Deleted broadcast 192.168.5.255 dev dummy1 table local proto kernel scope link src 192.168.5.5
Deleted local 192.168.5.5 dev dummy1 table local proto kernel scope host src 192.168.5.5
Deleted 7.7.7.0/24 dev dummy2 scope link src 192.168.5.5
Suggested-by: Thomas Haller <thaller@redhat.com>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20230922075508.848925-1-liuhangbin@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 8702cf12e6 upstream.
syzbot reported a warning [0] introduced by commit c48ef9c4ae ("tcp: Fix
bind() regression for v4-mapped-v6 non-wildcard address.").
After the cited commit, a v4 socket's address matches the corresponding
v4-mapped-v6 tb2 in inet_bind2_bucket_match_addr(), not vice versa.
During X.X.X.X -> ::ffff:X.X.X.X order bind()s, the second bind() uses
bhash and conflicts properly without checking bhash2 so that we need not
check if a v4-mapped-v6 sk matches the corresponding v4 address tb2 in
inet_bind2_bucket_match_addr(). However, the repro shows that we need
to check that in a no-conflict case.
The repro bind()s two sockets to the 2-tuples using SO_REUSEPORT and calls
listen() for the first socket:
from socket import *
s1 = socket()
s1.setsockopt(SOL_SOCKET, SO_REUSEPORT, 1)
s1.bind(('127.0.0.1', 0))
s2 = socket(AF_INET6)
s2.setsockopt(SOL_SOCKET, SO_REUSEPORT, 1)
s2.bind(('::ffff:127.0.0.1', s1.getsockname()[1]))
s1.listen()
The second socket should belong to the first socket's tb2, but the second
bind() creates another tb2 bucket because inet_bind2_bucket_find() returns
NULL in inet_csk_get_port() as the v4-mapped-v6 sk does not match the
corresponding v4 address tb2.
bhash2[] -> tb2(::ffff:X.X.X.X) -> tb2(X.X.X.X)
Then, listen() for the first socket calls inet_csk_get_port(), where the
v4 address matches the v4-mapped-v6 tb2 and WARN_ON() is triggered.
To avoid that, we need to check if v4-mapped-v6 sk address matches with
the corresponding v4 address tb2 in inet_bind2_bucket_match().
The same checks are needed in inet_bind2_bucket_addr_match() too, so we
can move all checks there and call it from inet_bind2_bucket_match().
Note that now tb->family is just an address family of tb->(v6_)?rcv_saddr
and not of sockets in the bucket. This could be refactored later by
defining tb->rcv_saddr as tb->v6_rcv_saddr.s6_addr32[3] and prepending
::ffff: when creating v4 tb2.
[0]:
WARNING: CPU: 0 PID: 5049 at net/ipv4/inet_connection_sock.c:587 inet_csk_get_port+0xf96/0x2350 net/ipv4/inet_connection_sock.c:587
Modules linked in:
CPU: 0 PID: 5049 Comm: syz-executor288 Not tainted 6.6.0-rc2-syzkaller-00018-g2cf0f7156238 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 08/04/2023
RIP: 0010:inet_csk_get_port+0xf96/0x2350 net/ipv4/inet_connection_sock.c:587
Code: 7c 24 08 e8 4c b6 8a 01 31 d2 be 88 01 00 00 48 c7 c7 e0 94 ae 8b e8 59 2e a3 f8 2e 2e 2e 31 c0 e9 04 fe ff ff e8 ca 88 d0 f8 <0f> 0b e9 0f f9 ff ff e8 be 88 d0 f8 49 8d 7e 48 e8 65 ca 5a 00 31
RSP: 0018:ffffc90003abfbf0 EFLAGS: 00010293
RAX: 0000000000000000 RBX: ffff888026429100 RCX: 0000000000000000
RDX: ffff88807edcbb80 RSI: ffffffff88b73d66 RDI: ffff888026c49f38
RBP: ffff888026c49f30 R08: 0000000000000005 R09: 0000000000000000
R10: 0000000000000001 R11: 0000000000000000 R12: ffffffff9260f200
R13: ffff888026c49880 R14: 0000000000000000 R15: ffff888026429100
FS: 00005555557d5380(0000) GS:ffff8880b9800000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 000000000045ad50 CR3: 0000000025754000 CR4: 00000000003506f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
inet_csk_listen_start+0x155/0x360 net/ipv4/inet_connection_sock.c:1256
__inet_listen_sk+0x1b8/0x5c0 net/ipv4/af_inet.c:217
inet_listen+0x93/0xd0 net/ipv4/af_inet.c:239
__sys_listen+0x194/0x270 net/socket.c:1866
__do_sys_listen net/socket.c:1875 [inline]
__se_sys_listen net/socket.c:1873 [inline]
__x64_sys_listen+0x53/0x80 net/socket.c:1873
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x38/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
RIP: 0033:0x7f3a5bce3af9
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 c1 17 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffc1a1c79e8 EFLAGS: 00000246 ORIG_RAX: 0000000000000032
RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f3a5bce3af9
RDX: 00007f3a5bce3af9 RSI: 0000000000000000 RDI: 0000000000000003
RBP: 00007f3a5bd565f0 R08: 0000000000000006 R09: 0000000000000006
R10: 0000000000000006 R11: 0000000000000246 R12: 0000000000000001
R13: 431bde82d7b634db R14: 0000000000000001 R15: 0000000000000001
</TASK>
Fixes: c48ef9c4ae ("tcp: Fix bind() regression for v4-mapped-v6 non-wildcard address.")
Reported-by: syzbot+71e724675ba3958edb31@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=71e724675ba3958edb31
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20231010013814.70571-1-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit f921a4a5bf upstream.
In commit 75eefc6c59 ("tcp: tsq: add a shortcut in tcp_small_queue_check()")
we allowed to send an skb regardless of TSQ limits being hit if rtx queue
was empty or had a single skb, in order to better fill the pipe
when/if TX completions were slow.
Then later, commit 75c119afe1 ("tcp: implement rb-tree based
retransmit queue") accidentally removed the special case for
one skb in rtx queue.
Stefan Wahren reported a regression in single TCP flow throughput
using a 100Mbit fec link, starting from commit 65466904b0 ("tcp: adjust
TSO packet sizes based on min_rtt"). This last commit only made the
regression more visible, because it locked the TCP flow on a particular
behavior where TSQ prevented two skbs being pushed downstream,
adding silences on the wire between each TSO packet.
Many thanks to Stefan for his invaluable help !
Fixes: 75c119afe1 ("tcp: implement rb-tree based retransmit queue")
Link: https://lore.kernel.org/netdev/7f31ddc8-9971-495e-a1f6-819df542e0af@gmx.net/
Reported-by: Stefan Wahren <wahrenst@gmx.net>
Tested-by: Stefan Wahren <wahrenst@gmx.net>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Link: https://lore.kernel.org/r/20231017124526.4060202-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 1c2709cfff upstream.
We discovered from packet traces of slow loss recovery on kernels with
the default HZ=250 setting (and min_rtt < 1ms) that after reordering,
when receiving a SACKed sequence range, the RACK reordering timer was
firing after about 16ms rather than the desired value of roughly
min_rtt/4 + 2ms. The problem is largely due to the RACK reorder timer
calculation adding in TCP_TIMEOUT_MIN, which is 2 jiffies. On kernels
with HZ=250, this is 2*4ms = 8ms. The TLP timer calculation has the
exact same issue.
This commit fixes the TLP transmit timer and RACK reordering timer
floor calculation to more closely match the intended 2ms floor even on
kernels with HZ=250. It does this by adding in a new
TCP_TIMEOUT_MIN_US floor of 2000 us and then converting to jiffies,
instead of the current approach of converting to jiffies and then
adding th TCP_TIMEOUT_MIN value of 2 jiffies.
Our testing has verified that on kernels with HZ=1000, as expected,
this does not produce significant changes in behavior, but on kernels
with the default HZ=250 the latency improvement can be large. For
example, our tests show that for HZ=250 kernels at low RTTs this fix
roughly halves the latency for the RACK reorder timer: instead of
mostly firing at 16ms it mostly fires at 8ms.
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Fixes: bb4d991a28 ("tcp: adjust tail loss probe timeout")
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20231015174700.2206872-1-ncardwell.sw@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 513f61e219 upstream.
In esp_remove_trailer(), to avoid an unexpected result returned by
pskb_trim, we should check the return value of pskb_trim().
Signed-off-by: Ma Ke <make_ruc2021@163.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 6db8a37dfc upstream.
The MPTCP protocol can acquire the subflow-level socket lock and
cause the tcp backlog usage. When inserting new skbs into the
backlog, the stack will try to coalesce them.
Currently, we have no check in place to ensure that such coalescing
will respect the MPTCP-level DSS, and that may cause data stream
corruption, as reported by Christoph.
Address the issue by adding the relevant admission check for coalescing
in tcp_add_backlog().
Note the issue is not easy to reproduce, as the MPTCP protocol tries
hard to avoid acquiring the subflow-level socket lock.
Fixes: 648ef4b886 ("mptcp: Implement MPTCP receive path")
Cc: stable@vger.kernel.org
Reported-by: Christoph Paasch <cpaasch@apple.com>
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/420
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <martineau@kernel.org>
Link: https://lore.kernel.org/r/20231018-send-net-20231018-v1-2-17ecb002e41d@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit b650d953cd ]
Under certain circumstances, the tcp receive buffer memory limit
set by autotuning (sk_rcvbuf) is increased due to incoming data
packets as a result of the window not closing when it should be.
This can result in the receive buffer growing all the way up to
tcp_rmem[2], even for tcp sessions with a low BDP.
To reproduce: Connect a TCP session with the receiver doing
nothing and the sender sending small packets (an infinite loop
of socket send() with 4 bytes of payload with a sleep of 1 ms
in between each send()). This will cause the tcp receive buffer
to grow all the way up to tcp_rmem[2].
As a result, a host can have individual tcp sessions with receive
buffers of size tcp_rmem[2], and the host itself can reach tcp_mem
limits, causing the host to go into tcp memory pressure mode.
The fundamental issue is the relationship between the granularity
of the window scaling factor and the number of byte ACKed back
to the sender. This problem has previously been identified in
RFC 7323, appendix F [1].
The Linux kernel currently adheres to never shrinking the window.
In addition to the overallocation of memory mentioned above, the
current behavior is functionally incorrect, because once tcp_rmem[2]
is reached when no remediations remain (i.e. tcp collapse fails to
free up any more memory and there are no packets to prune from the
out-of-order queue), the receiver will drop in-window packets
resulting in retransmissions and an eventual timeout of the tcp
session. A receive buffer full condition should instead result
in a zero window and an indefinite wait.
In practice, this problem is largely hidden for most flows. It
is not applicable to mice flows. Elephant flows can send data
fast enough to "overrun" the sk_rcvbuf limit (in a single ACK),
triggering a zero window.
But this problem does show up for other types of flows. Examples
are websockets and other type of flows that send small amounts of
data spaced apart slightly in time. In these cases, we directly
encounter the problem described in [1].
RFC 7323, section 2.4 [2], says there are instances when a retracted
window can be offered, and that TCP implementations MUST ensure
that they handle a shrinking window, as specified in RFC 1122,
section 4.2.2.16 [3]. All prior RFCs on the topic of tcp window
management have made clear that sender must accept a shrunk window
from the receiver, including RFC 793 [4] and RFC 1323 [5].
This patch implements the functionality to shrink the tcp window
when necessary to keep the right edge within the memory limit by
autotuning (sk_rcvbuf). This new functionality is enabled with
the new sysctl: net.ipv4.tcp_shrink_window
Additional information can be found at:
https://blog.cloudflare.com/unbounded-memory-usage-by-tcp-for-receive-buffers-and-how-we-fixed-it/
[1] https://www.rfc-editor.org/rfc/rfc7323#appendix-F
[2] https://www.rfc-editor.org/rfc/rfc7323#section-2.4
[3] https://www.rfc-editor.org/rfc/rfc1122#page-91
[4] https://www.rfc-editor.org/rfc/rfc793
[5] https://www.rfc-editor.org/rfc/rfc1323
Signed-off-by: Mike Freemon <mfreemon@cloudflare.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 4720852ed9 ]
This commit fixes poor delayed ACK behavior that can cause poor TCP
latency in a particular boundary condition: when an application makes
a TCP socket write that is an exact multiple of the MSS size.
The problem is that there is painful boundary discontinuity in the
current delayed ACK behavior. With the current delayed ACK behavior,
we have:
(1) If an app reads data when > 1*MSS is unacknowledged, then
tcp_cleanup_rbuf() ACKs immediately because of:
tp->rcv_nxt - tp->rcv_wup > icsk->icsk_ack.rcv_mss ||
(2) If an app reads all received data, and the packets were < 1*MSS,
and either (a) the app is not ping-pong or (b) we received two
packets < 1*MSS, then tcp_cleanup_rbuf() ACKs immediately beecause
of:
((icsk->icsk_ack.pending & ICSK_ACK_PUSHED2) ||
((icsk->icsk_ack.pending & ICSK_ACK_PUSHED) &&
!inet_csk_in_pingpong_mode(sk))) &&
(3) *However*: if an app reads exactly 1*MSS of data,
tcp_cleanup_rbuf() does not send an immediate ACK. This is true
even if the app is not ping-pong and the 1*MSS of data had the PSH
bit set, suggesting the sending application completed an
application write.
Thus if the app is not ping-pong, we have this painful case where
>1*MSS gets an immediate ACK, and <1*MSS gets an immediate ACK, but a
write whose last skb is an exact multiple of 1*MSS can get a 40ms
delayed ACK. This means that any app that transfers data in one
direction and takes care to align write size or packet size with MSS
can suffer this problem. With receive zero copy making 4KB MSS values
more common, it is becoming more common to have application writes
naturally align with MSS, and more applications are likely to
encounter this delayed ACK problem.
The fix in this commit is to refine the delayed ACK heuristics with a
simple check: immediately ACK a received 1*MSS skb with PSH bit set if
the app reads all data. Why? If an skb has a len of exactly 1*MSS and
has the PSH bit set then it is likely the end of an application
write. So more data may not be arriving soon, and yet the data sender
may be waiting for an ACK if cwnd-bound or using TX zero copy. Thus we
set ICSK_ACK_PUSHED in this case so that tcp_cleanup_rbuf() will send
an ACK immediately if the app reads all of the data and is not
ping-pong. Note that this logic is also executed for the case where
len > MSS, but in that case this logic does not matter (and does not
hurt) because tcp_cleanup_rbuf() will always ACK immediately if the
app reads data and there is more than an MSS of unACKed data.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Yuchung Cheng <ycheng@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Cc: Xin Guo <guoxin0309@gmail.com>
Link: https://lore.kernel.org/r/20231001151239.1866845-2-ncardwell.sw@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 059217c18b ]
This commit fixes quick-ack counting so that it only considers that a
quick-ack has been provided if we are sending an ACK that newly
acknowledges data.
The code was erroneously using the number of data segments in outgoing
skbs when deciding how many quick-ack credits to remove. This logic
does not make sense, and could cause poor performance in
request-response workloads, like RPC traffic, where requests or
responses can be multi-segment skbs.
When a TCP connection decides to send N quick-acks, that is to
accelerate the cwnd growth of the congestion control module
controlling the remote endpoint of the TCP connection. That quick-ack
decision is purely about the incoming data and outgoing ACKs. It has
nothing to do with the outgoing data or the size of outgoing data.
And in particular, an ACK only serves the intended purpose of allowing
the remote congestion control to grow the congestion window quickly if
the ACK is ACKing or SACKing new data.
The fix is simple: only count packets as serving the goal of the
quickack mechanism if they are ACKing/SACKing new data. We can tell
whether this is the case by checking inet_csk_ack_scheduled(), since
we schedule an ACK exactly when we are ACKing/SACKing new data.
Fixes: fc6415bcb0 ("[TCP]: Fix quick-ack decrementing with TSO.")
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Yuchung Cheng <ycheng@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20231001151239.1866845-1-ncardwell.sw@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 0add5c597f ]
Due to a small omission, the offload_failed flag is missing from ipv4
fibmatch results. Make sure it is set correctly.
The issue can be witnessed using the following commands:
echo "1 1" > /sys/bus/netdevsim/new_device
ip link add dummy1 up type dummy
ip route add 192.0.2.0/24 dev dummy1
echo 1 > /sys/kernel/debug/netdevsim/netdevsim1/fib/fail_route_offload
ip route add 198.51.100.0/24 dev dummy1
ip route
# 192.168.15.0/24 has rt_trap
# 198.51.100.0/24 has rt_offload_failed
ip route get 192.168.15.1 fibmatch
# Result has rt_trap
ip route get 198.51.100.1 fibmatch
# Result differs from the route shown by `ip route`, it is missing
# rt_offload_failed
ip link del dev dummy1
echo 1 > /sys/bus/netdevsim/del_device
Fixes: 36c5100e85 ("IPv4: Add "offload failed" indication to routes")
Signed-off-by: Benjamin Poirier <bpoirier@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20230926182730.231208-1-bpoirier@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 09eed1192c ]
rcu_bh is no longer a win, especially for objects freed
with standard call_rcu().
Switch neighbour code to no longer disable BH when not necessary.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Stable-dep-of: 5baa0433a1 ("neighbour: fix data-races around n->output")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit b071af5235 ]
We have many lockless accesses to n->nud_state.
Before adding another one in the following patch,
add annotations to readers and writers.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Stable-dep-of: 5baa0433a1 ("neighbour: fix data-races around n->output")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit da9e915eaf ]
When data is peek'd off the receive queue we shouldn't considered it
copied from tcp_sock side. When we increment copied_seq this will confuse
tcp_data_ready() because copied_seq can be arbitrarily increased. From
application side it results in poll() operations not waking up when
expected.
Notice tcp stack without BPF recvmsg programs also does not increment
copied_seq.
We broke this when we moved copied_seq into recvmsg to only update when
actual copy was happening. But, it wasn't working correctly either before
because the tcp_data_ready() tried to use the copied_seq value to see
if data was read by user yet. See fixes tags.
Fixes: e5c6de5fa0 ("bpf, sockmap: Incorrectly handling copied_seq")
Fixes: 04919bed94 ("tcp: Introduce tcp_read_skb()")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Link: https://lore.kernel.org/bpf/20230926035300.135096-3-john.fastabend@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 9b7177b1df ]
Before fix e5c6de5fa0 tcp_read_skb() would increment the tp->copied-seq
value. This (as described in the commit) would cause an error for apps
because once that is incremented the application might believe there is no
data to be read. Then some apps would stall or abort believing no data is
available.
However, the fix is incomplete because it introduces another issue in
the skb dequeue. The loop does tcp_recv_skb() in a while loop to consume
as many skbs as possible. The problem is the call is ...
tcp_recv_skb(sk, seq, &offset)
... where 'seq' is:
u32 seq = tp->copied_seq;
Now we can hit a case where we've yet incremented copied_seq from BPF side,
but then tcp_recv_skb() fails this test ...
if (offset < skb->len || (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN))
... so that instead of returning the skb we call tcp_eat_recv_skb() which
frees the skb. This is because the routine believes the SKB has been collapsed
per comment:
/* This looks weird, but this can happen if TCP collapsing
* splitted a fat GRO packet, while we released socket lock
* in skb_splice_bits()
*/
This can't happen here we've unlinked the full SKB and orphaned it. Anyways
it would confuse any BPF programs if the data were suddenly moved underneath
it.
To fix this situation do simpler operation and just skb_peek() the data
of the queue followed by the unlink. It shouldn't need to check this
condition and tcp_read_skb() reads entire skbs so there is no need to
handle the 'offset!=0' case as we would see in tcp_read_sock().
Fixes: e5c6de5fa0 ("bpf, sockmap: Incorrectly handling copied_seq")
Fixes: 04919bed94 ("tcp: Introduce tcp_read_skb()")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Link: https://lore.kernel.org/bpf/20230926035300.135096-2-john.fastabend@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 0113d9c9d1 ]
Currently, we assume the skb is associated with a device before calling
__ip_options_compile, which is not always the case if it is re-routed by
ipvs.
When skb->dev is NULL, dev_net(skb->dev) will become null-dereference.
This patch adds a check for the edge case and switch to use the net_device
from the rtable when skb->dev is NULL.
Fixes: ed0de45a10 ("ipv4: recompile ip options in ipv4_link_failure")
Suggested-by: David Ahern <dsahern@kernel.org>
Signed-off-by: Kyle Zeng <zengyhkyle@gmail.com>
Cc: Stephen Suryaputra <ssuryaextr@gmail.com>
Cc: Vadim Fedorenko <vfedorenko@novek.ru>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit c67180efc5 ]
For now, No matter what error pointer ip_neigh_for_gw() returns,
ip_finish_output2() always return -EINVAL, which may mislead the upper
users.
For exemple, an application uses sendto to send an UDP packet, but when the
neighbor table overflows, sendto() will get a value of -EINVAL, and it will
cause users to waste a lot of time checking parameters for errors.
Return the real errno instead of -EINVAL.
Signed-off-by: xu xin <xu.xin16@zte.com.cn>
Reviewed-by: Yang Yang <yang.yang29@zte.com.cn>
Cc: Si Hao <si.hao@zte.com.cn>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
Link: https://lore.kernel.org/r/20230807015408.248237-1-xu.xin16@zte.com.cn
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit c48ef9c4ae ]
Since bhash2 was introduced, the example below does not work as expected.
These two bind() should conflict, but the 2nd bind() now succeeds.
from socket import *
s1 = socket(AF_INET6, SOCK_STREAM)
s1.bind(('::ffff:127.0.0.1', 0))
s2 = socket(AF_INET, SOCK_STREAM)
s2.bind(('127.0.0.1', s1.getsockname()[1]))
During the 2nd bind() in inet_csk_get_port(), inet_bind2_bucket_find()
fails to find the 1st socket's tb2, so inet_bind2_bucket_create() allocates
a new tb2 for the 2nd socket. Then, we call inet_csk_bind_conflict() that
checks conflicts in the new tb2 by inet_bhash2_conflict(). However, the
new tb2 does not include the 1st socket, thus the bind() finally succeeds.
In this case, inet_bind2_bucket_match() must check if AF_INET6 tb2 has
the conflicting v4-mapped-v6 address so that inet_bind2_bucket_find()
returns the 1st socket's tb2.
Note that if we bind two sockets to 127.0.0.1 and then ::FFFF:127.0.0.1,
the 2nd bind() fails properly for the same reason mentinoed in the previous
commit.
Fixes: 28044fc1d4 ("net: Add a bhash2 table hashed by port and address")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: Andrei Vagin <avagin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit aa99e5f87b ]
Andrei Vagin reported bind() regression with strace logs.
If we bind() a TCPv6 socket to ::FFFF:0.0.0.0 and then bind() a TCPv4
socket to 127.0.0.1, the 2nd bind() should fail but now succeeds.
from socket import *
s1 = socket(AF_INET6, SOCK_STREAM)
s1.bind(('::ffff:0.0.0.0', 0))
s2 = socket(AF_INET, SOCK_STREAM)
s2.bind(('127.0.0.1', s1.getsockname()[1]))
During the 2nd bind(), if tb->family is AF_INET6 and sk->sk_family is
AF_INET in inet_bind2_bucket_match_addr_any(), we still need to check
if tb has the v4-mapped-v6 wildcard address.
The example above does not work after commit 5456262d2b ("net: Fix
incorrect address comparison when searching for a bind2 bucket"), but
the blamed change is not the commit.
Before the commit, the leading zeros of ::FFFF:0.0.0.0 were treated
as 0.0.0.0, and the sequence above worked by chance. Technically, this
case has been broken since bhash2 was introduced.
Note that if we bind() two sockets to 127.0.0.1 and then ::FFFF:0.0.0.0,
the 2nd bind() fails properly because we fall back to using bhash to
detect conflicts for the v4-mapped-v6 address.
Fixes: 28044fc1d4 ("net: Add a bhash2 table hashed by port and address")
Reported-by: Andrei Vagin <avagin@google.com>
Closes: https://lore.kernel.org/netdev/ZPuYBOFC8zsK6r9T@google.com/
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit c6d277064b ]
This is a prep patch to make the following patches cleaner that touch
inet_bind2_bucket_match() and inet_bind2_bucket_match_addr_any().
Both functions have duplicated comparison for netns, port, and l3mdev.
Let's factorise them.
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stable-dep-of: aa99e5f87b ("tcp: Fix bind() regression for v4-mapped-v6 wildcard address.")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 8cdc3223e7 ]
Some code defines the IPv6 wildcard address as a local variable and
use it with memcmp() or ipv6_addr_equal().
Let's use in6addr_any and ipv6_addr_any() instead.
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stable-dep-of: aa99e5f87b ("tcp: Fix bind() regression for v4-mapped-v6 wildcard address.")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit ac28b1ec61 ]
I got the below warning when do fuzzing test:
unregister_netdevice: waiting for bond0 to become free. Usage count = 2
It can be repoduced via:
ip link add bond0 type bond
sysctl -w net.ipv4.conf.bond0.promote_secondaries=1
ip addr add 4.117.174.103/0 scope 0x40 dev bond0
ip addr add 192.168.100.111/255.255.255.254 scope 0 dev bond0
ip addr add 0.0.0.4/0 scope 0x40 secondary dev bond0
ip addr del 4.117.174.103/0 scope 0x40 dev bond0
ip link delete bond0 type bond
In this reproduction test case, an incorrect 'last_prim' is found in
__inet_del_ifa(), as a result, the secondary address(0.0.0.4/0 scope 0x40)
is lost. The memory of the secondary address is leaked and the reference of
in_device and net_device is leaked.
Fix this problem:
Look for 'last_prim' starting at location of the deleted IP and inserting
the promoted IP into the location of 'last_prim'.
Fixes: 0ff60a4567 ("[IPV4]: Fix secondary IP addresses after promotion")
Signed-off-by: Liu Jian <liujian56@huawei.com>
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 6ac66cb03a ]
Route hints when the nexthop is part of a multipath group causes packets
in the same receive batch to be sent to the same nexthop irrespective of
the multipath hash of the packet. So, do not extract route hint for
packets whose destination is part of a multipath group.
A new SKB flag IPSKB_MULTIPATH is introduced for this purpose, set the
flag when route is looked up in ip_mkroute_input() and use it in
ip_extract_route_hint() to check for the existence of the flag.
Fixes: 02b2494161 ("ipv4: use dst hint for ipv4 list receive")
Signed-off-by: Sriram Yagnaraman <sriram.yagnaraman@est.tech>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 5e6300e7b3 ]
Every time sk->sk_forward_alloc is read locklessly,
add a READ_ONCE().
Add sk_forward_alloc_add() helper to centralize updates,
to reduce number of WRITE_ONCE().
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit c3b704d4a4 upstream.
This is a follow up of commit 915d975b2f ("net: deal with integer
overflows in kmalloc_reserve()") based on David Laight feedback.
Back in 2010, I failed to realize malicious users could set dev->mtu
to arbitrary values. This mtu has been since limited to 0x7fffffff but
regardless of how big dev->mtu is, it makes no sense for igmpv3_newpack()
to allocate more than IP_MAX_MTU and risk various skb fields overflows.
Fixes: 57e1ab6ead ("igmp: refine skb allocations")
Link: https://lore.kernel.org/netdev/d273628df80f45428e739274ab9ecb72@AcuMS.aculab.com/
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: David Laight <David.Laight@ACULAB.COM>
Cc: Kyle Zeng <zengyhkyle@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit a171fbec88 ]
LWTUNNEL_XMIT_CONTINUE is implicitly assumed in ip(6)_finish_output2,
such that any positive return value from a xmit hook could cause
unexpected continue behavior, despite that related skb may have been
freed. This could be error-prone for future xmit hook ops. One of the
possible errors is to return statuses of dst_output directly.
To make the code safer, redefine LWTUNNEL_XMIT_CONTINUE value to
distinguish from dst_output statuses and check the continue
condition explicitly.
Fixes: 3a0af8fd61 ("bpf: BPF for lightweight tunnel infrastructure")
Suggested-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Yan Zhai <yan@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/96b939b85eda00e8df4f7c080f770970a4c5f698.1692326837.git.yan@cloudflare.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e89688e3e9 ]
In tcp_retransmit_timer(), a window shrunk connection will be regarded
as timeout if 'tcp_jiffies32 - tp->rcv_tstamp > TCP_RTO_MAX'. This is not
right all the time.
The retransmits will become zero-window probes in tcp_retransmit_timer()
if the 'snd_wnd==0'. Therefore, the icsk->icsk_rto will come up to
TCP_RTO_MAX sooner or later.
However, the timer can be delayed and be triggered after 122877ms, not
TCP_RTO_MAX, as I tested.
Therefore, 'tcp_jiffies32 - tp->rcv_tstamp > TCP_RTO_MAX' is always true
once the RTO come up to TCP_RTO_MAX, and the socket will die.
Fix this by replacing the 'tcp_jiffies32' with '(u32)icsk->icsk_timeout',
which is exact the timestamp of the timeout.
However, "tp->rcv_tstamp" can restart from idle, then tp->rcv_tstamp
could already be a long time (minutes or hours) in the past even on the
first RTO. So we double check the timeout with the duration of the
retransmission.
Meanwhile, making "2 * TCP_RTO_MAX" as the timeout to avoid the socket
dying too soon.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Link: https://lore.kernel.org/netdev/CADxym3YyMiO+zMD4zj03YPM3FBi-1LHi6gSD2XT8pyAMM096pg@mail.gmail.com/
Signed-off-by: Menglong Dong <imagedong@tencent.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit f0ea27e7bf ]
Contrary to TCP, UDP reuseport groups can contain TCP_ESTABLISHED
sockets. To support these properly we remember whether a group has
a connected socket and skip the fast reuseport early-return. In
effect we continue scoring all reuseport sockets and then choose the
one with the highest score.
The current code fails to re-calculate the score for the result of
lookup_reuseport. According to Kuniyuki Iwashima:
1) SO_INCOMING_CPU is set
-> selected sk might have +1 score
2) BPF prog returns ESTABLISHED and/or SO_INCOMING_CPU sk
-> selected sk will have more than 8
Using the old score could trigger more lookups depending on the
order that sockets are created.
sk -> sk (SO_INCOMING_CPU) -> sk (ESTABLISHED)
| |
`-> select the next SO_INCOMING_CPU sk
|
`-> select itself (We should save this lookup)
Fixes: efc6b6f6c3 ("udp: Improve load balancing for SO_REUSEPORT.")
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Lorenz Bauer <lmb@isovalent.com>
Link: https://lore.kernel.org/r/20230720-so-reuseport-v6-1-7021b683cdae@isovalent.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit f866fbc842 ]
UDP sendmsg() is lockless, so ip_select_ident_segs()
can very well be run from multiple cpus [1]
Convert inet->inet_id to an atomic_t, but implement
a dedicated path for TCP, avoiding cost of a locked
instruction (atomic_add_return())
Note that this patch will cause a trivial merge conflict
because we added inet->flags in net-next tree.
v2: added missing change in
drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c
(David Ahern)
[1]
BUG: KCSAN: data-race in __ip_make_skb / __ip_make_skb
read-write to 0xffff888145af952a of 2 bytes by task 7803 on cpu 1:
ip_select_ident_segs include/net/ip.h:542 [inline]
ip_select_ident include/net/ip.h:556 [inline]
__ip_make_skb+0x844/0xc70 net/ipv4/ip_output.c:1446
ip_make_skb+0x233/0x2c0 net/ipv4/ip_output.c:1560
udp_sendmsg+0x1199/0x1250 net/ipv4/udp.c:1260
inet_sendmsg+0x63/0x80 net/ipv4/af_inet.c:830
sock_sendmsg_nosec net/socket.c:725 [inline]
sock_sendmsg net/socket.c:748 [inline]
____sys_sendmsg+0x37c/0x4d0 net/socket.c:2494
___sys_sendmsg net/socket.c:2548 [inline]
__sys_sendmmsg+0x269/0x500 net/socket.c:2634
__do_sys_sendmmsg net/socket.c:2663 [inline]
__se_sys_sendmmsg net/socket.c:2660 [inline]
__x64_sys_sendmmsg+0x57/0x60 net/socket.c:2660
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
read to 0xffff888145af952a of 2 bytes by task 7804 on cpu 0:
ip_select_ident_segs include/net/ip.h:541 [inline]
ip_select_ident include/net/ip.h:556 [inline]
__ip_make_skb+0x817/0xc70 net/ipv4/ip_output.c:1446
ip_make_skb+0x233/0x2c0 net/ipv4/ip_output.c:1560
udp_sendmsg+0x1199/0x1250 net/ipv4/udp.c:1260
inet_sendmsg+0x63/0x80 net/ipv4/af_inet.c:830
sock_sendmsg_nosec net/socket.c:725 [inline]
sock_sendmsg net/socket.c:748 [inline]
____sys_sendmsg+0x37c/0x4d0 net/socket.c:2494
___sys_sendmsg net/socket.c:2548 [inline]
__sys_sendmmsg+0x269/0x500 net/socket.c:2634
__do_sys_sendmmsg net/socket.c:2663 [inline]
__se_sys_sendmmsg net/socket.c:2660 [inline]
__x64_sys_sendmmsg+0x57/0x60 net/socket.c:2660
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
value changed: 0x184d -> 0x184e
Reported by Kernel Concurrency Sanitizer on:
CPU: 0 PID: 7804 Comm: syz-executor.1 Not tainted 6.5.0-rc6-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/26/2023
==================================================================
Fixes: 23f57406b8 ("ipv4: avoid using shared IP generator for connected sockets")
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit e4dd0d3a2f upstream.
In the real workload, I encountered an issue which could cause the RTO
timer to retransmit the skb per 1ms with linear option enabled. The amount
of lost-retransmitted skbs can go up to 1000+ instantly.
The root cause is that if the icsk_rto happens to be zero in the 6th round
(which is the TCP_THIN_LINEAR_RETRIES value), then it will always be zero
due to the changed calculation method in tcp_retransmit_timer() as follows:
icsk->icsk_rto = min(icsk->icsk_rto << 1, TCP_RTO_MAX);
Above line could be converted to
icsk->icsk_rto = min(0 << 1, TCP_RTO_MAX) = 0
Therefore, the timer expires so quickly without any doubt.
I read through the RFC 6298 and found that the RTO value can be rounded
up to a certain value, in Linux, say TCP_RTO_MIN as default, which is
regarded as the lower bound in this patch as suggested by Eric.
Fixes: 36e31b0af5 ("net: TCP thin linear timeouts")
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jason Xing <kernelxing@tencent.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 6018a26627 ]
When ip_vti device is set to the qdisc of the sfb type, the cb field
of the sent skb may be modified during enqueuing. Then,
slab-use-after-free may occur when ip_vti device sends IPv6 packets.
As commit f855691975 ("xfrm6: Fix the nexthdr offset in
_decode_session6.") showed, xfrm_decode_session was originally intended
only for the receive path. IP6CB(skb)->nhoff is not set during
transmission. Therefore, set the cb field in the skb to 0 before
sending packets.
Fixes: f855691975 ("xfrm6: Fix the nexthdr offset in _decode_session6.")
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 8743aeff5b upstream.
A netlink dump callback can return a positive number to signal that more
information needs to be dumped or zero to signal that the dump is
complete. In the second case, the core netlink code will append the
NLMSG_DONE message to the skb in order to indicate to user space that
the dump is complete.
The nexthop bucket dump callback always returns a positive number if
nexthop buckets were filled in the provided skb, even if the dump is
complete. This means that a dump will span at least two recvmsg() calls
as long as nexthop buckets are present. In the last recvmsg() call the
dump callback will not fill in any nexthop buckets because the previous
call indicated that the dump should restart from the last dumped nexthop
ID plus one.
# ip link add name dummy1 up type dummy
# ip nexthop add id 1 dev dummy1
# ip nexthop add id 10 group 1 type resilient buckets 2
# strace -e sendto,recvmsg -s 5 ip nexthop bucket
sendto(3, [[{nlmsg_len=24, nlmsg_type=RTM_GETNEXTHOPBUCKET, nlmsg_flags=NLM_F_REQUEST|NLM_F_DUMP, nlmsg_seq=1691396980, nlmsg_pid=0}, {family=AF_UNSPEC, data="\x00\x00\x00\x00\x00"...}], {nlmsg_len=0, nlmsg_type=0 /* NLMSG_??? */, nlmsg_flags=0, nlmsg_seq=0, nlmsg_pid=0}], 152, 0, NULL, 0) = 152
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[{iov_base=NULL, iov_len=0}], msg_iovlen=1, msg_controllen=0, msg_flags=MSG_TRUNC}, MSG_PEEK|MSG_TRUNC) = 128
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[{iov_base=[[{nlmsg_len=64, nlmsg_type=RTM_NEWNEXTHOPBUCKET, nlmsg_flags=NLM_F_MULTI, nlmsg_seq=1691396980, nlmsg_pid=347}, {family=AF_UNSPEC, data="\x00\x00\x00\x00\x00"...}], [{nlmsg_len=64, nlmsg_type=RTM_NEWNEXTHOPBUCKET, nlmsg_flags=NLM_F_MULTI, nlmsg_seq=1691396980, nlmsg_pid=347}, {family=AF_UNSPEC, data="\x00\x00\x00\x00\x00"...}]], iov_len=32768}], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 128
id 10 index 0 idle_time 6.66 nhid 1
id 10 index 1 idle_time 6.66 nhid 1
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[{iov_base=NULL, iov_len=0}], msg_iovlen=1, msg_controllen=0, msg_flags=MSG_TRUNC}, MSG_PEEK|MSG_TRUNC) = 20
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[{iov_base=[{nlmsg_len=20, nlmsg_type=NLMSG_DONE, nlmsg_flags=NLM_F_MULTI, nlmsg_seq=1691396980, nlmsg_pid=347}, 0], iov_len=32768}], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 20
+++ exited with 0 +++
This behavior is both inefficient and buggy. If the last nexthop to be
dumped had the maximum ID of 0xffffffff, then the dump will restart from
0 (0xffffffff + 1) and never end:
# ip link add name dummy1 up type dummy
# ip nexthop add id 1 dev dummy1
# ip nexthop add id $((2**32-1)) group 1 type resilient buckets 2
# ip nexthop bucket
id 4294967295 index 0 idle_time 5.55 nhid 1
id 4294967295 index 1 idle_time 5.55 nhid 1
id 4294967295 index 0 idle_time 5.55 nhid 1
id 4294967295 index 1 idle_time 5.55 nhid 1
[...]
Fix by adjusting the dump callback to return zero when the dump is
complete. After the fix only one recvmsg() call is made and the
NLMSG_DONE message is appended to the RTM_NEWNEXTHOPBUCKET responses:
# ip link add name dummy1 up type dummy
# ip nexthop add id 1 dev dummy1
# ip nexthop add id $((2**32-1)) group 1 type resilient buckets 2
# strace -e sendto,recvmsg -s 5 ip nexthop bucket
sendto(3, [[{nlmsg_len=24, nlmsg_type=RTM_GETNEXTHOPBUCKET, nlmsg_flags=NLM_F_REQUEST|NLM_F_DUMP, nlmsg_seq=1691396737, nlmsg_pid=0}, {family=AF_UNSPEC, data="\x00\x00\x00\x00\x00"...}], {nlmsg_len=0, nlmsg_type=0 /* NLMSG_??? */, nlmsg_flags=0, nlmsg_seq=0, nlmsg_pid=0}], 152, 0, NULL, 0) = 152
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[{iov_base=NULL, iov_len=0}], msg_iovlen=1, msg_controllen=0, msg_flags=MSG_TRUNC}, MSG_PEEK|MSG_TRUNC) = 148
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[{iov_base=[[{nlmsg_len=64, nlmsg_type=RTM_NEWNEXTHOPBUCKET, nlmsg_flags=NLM_F_MULTI, nlmsg_seq=1691396737, nlmsg_pid=350}, {family=AF_UNSPEC, data="\x00\x00\x00\x00\x00"...}], [{nlmsg_len=64, nlmsg_type=RTM_NEWNEXTHOPBUCKET, nlmsg_flags=NLM_F_MULTI, nlmsg_seq=1691396737, nlmsg_pid=350}, {family=AF_UNSPEC, data="\x00\x00\x00\x00\x00"...}], [{nlmsg_len=20, nlmsg_type=NLMSG_DONE, nlmsg_flags=NLM_F_MULTI, nlmsg_seq=1691396737, nlmsg_pid=350}, 0]], iov_len=32768}], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 148
id 4294967295 index 0 idle_time 6.61 nhid 1
id 4294967295 index 1 idle_time 6.61 nhid 1
+++ exited with 0 +++
Note that if the NLMSG_DONE message cannot be appended because of size
limitations, then another recvmsg() will be needed, but the core netlink
code will not invoke the dump callback and simply reply with a
NLMSG_DONE message since it knows that the callback previously returned
zero.
Add a test that fails before the fix:
# ./fib_nexthops.sh -t basic_res
[...]
TEST: Maximum nexthop ID dump [FAIL]
[...]
And passes after it:
# ./fib_nexthops.sh -t basic_res
[...]
TEST: Maximum nexthop ID dump [ OK ]
[...]
Fixes: 8a1bbabb03 ("nexthop: Add netlink handlers for bucket dump")
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20230808075233.3337922-4-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit f10d3d9df4 upstream.
rtm_dump_nexthop_bucket_nh() is used to dump nexthop buckets belonging
to a specific resilient nexthop group. The function returns a positive
return code (the skb length) upon both success and failure.
The above behavior is problematic. When a complete nexthop bucket dump
is requested, the function that walks the different nexthops treats the
non-zero return code as an error. This causes buckets belonging to
different resilient nexthop groups to be dumped using different buffers
even if they can all fit in the same buffer:
# ip link add name dummy1 up type dummy
# ip nexthop add id 1 dev dummy1
# ip nexthop add id 10 group 1 type resilient buckets 1
# ip nexthop add id 20 group 1 type resilient buckets 1
# strace -e recvmsg -s 0 ip nexthop bucket
[...]
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[...], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 64
id 10 index 0 idle_time 10.27 nhid 1
[...]
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[...], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 64
id 20 index 0 idle_time 6.44 nhid 1
[...]
Fix by only returning a non-zero return code when an error occurred and
restarting the dump from the bucket index we failed to fill in. This
allows buckets belonging to different resilient nexthop groups to be
dumped using the same buffer:
# ip link add name dummy1 up type dummy
# ip nexthop add id 1 dev dummy1
# ip nexthop add id 10 group 1 type resilient buckets 1
# ip nexthop add id 20 group 1 type resilient buckets 1
# strace -e recvmsg -s 0 ip nexthop bucket
[...]
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[...], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 128
id 10 index 0 idle_time 30.21 nhid 1
id 20 index 0 idle_time 26.7 nhid 1
[...]
While this change is more of a performance improvement change than an
actual bug fix, it is a prerequisite for a subsequent patch that does
fix a bug.
Fixes: 8a1bbabb03 ("nexthop: Add netlink handlers for bucket dump")
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20230808075233.3337922-3-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 913f60cacd upstream.
A netlink dump callback can return a positive number to signal that more
information needs to be dumped or zero to signal that the dump is
complete. In the second case, the core netlink code will append the
NLMSG_DONE message to the skb in order to indicate to user space that
the dump is complete.
The nexthop dump callback always returns a positive number if nexthops
were filled in the provided skb, even if the dump is complete. This
means that a dump will span at least two recvmsg() calls as long as
nexthops are present. In the last recvmsg() call the dump callback will
not fill in any nexthops because the previous call indicated that the
dump should restart from the last dumped nexthop ID plus one.
# ip nexthop add id 1 blackhole
# strace -e sendto,recvmsg -s 5 ip nexthop
sendto(3, [[{nlmsg_len=24, nlmsg_type=RTM_GETNEXTHOP, nlmsg_flags=NLM_F_REQUEST|NLM_F_DUMP, nlmsg_seq=1691394315, nlmsg_pid=0}, {nh_family=AF_UNSPEC, nh_scope=RT_SCOPE_UNIVERSE, nh_protocol=RTPROT_UNSPEC, nh_flags=0}], {nlmsg_len=0, nlmsg_type=0 /* NLMSG_??? */, nlmsg_flags=0, nlmsg_seq=0, nlmsg_pid=0}], 152, 0, NULL, 0) = 152
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[{iov_base=NULL, iov_len=0}], msg_iovlen=1, msg_controllen=0, msg_flags=MSG_TRUNC}, MSG_PEEK|MSG_TRUNC) = 36
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[{iov_base=[{nlmsg_len=36, nlmsg_type=RTM_NEWNEXTHOP, nlmsg_flags=NLM_F_MULTI, nlmsg_seq=1691394315, nlmsg_pid=343}, {nh_family=AF_INET, nh_scope=RT_SCOPE_UNIVERSE, nh_protocol=RTPROT_UNSPEC, nh_flags=0}, [[{nla_len=8, nla_type=NHA_ID}, 1], {nla_len=4, nla_type=NHA_BLACKHOLE}]], iov_len=32768}], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 36
id 1 blackhole
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[{iov_base=NULL, iov_len=0}], msg_iovlen=1, msg_controllen=0, msg_flags=MSG_TRUNC}, MSG_PEEK|MSG_TRUNC) = 20
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[{iov_base=[{nlmsg_len=20, nlmsg_type=NLMSG_DONE, nlmsg_flags=NLM_F_MULTI, nlmsg_seq=1691394315, nlmsg_pid=343}, 0], iov_len=32768}], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 20
+++ exited with 0 +++
This behavior is both inefficient and buggy. If the last nexthop to be
dumped had the maximum ID of 0xffffffff, then the dump will restart from
0 (0xffffffff + 1) and never end:
# ip nexthop add id $((2**32-1)) blackhole
# ip nexthop
id 4294967295 blackhole
id 4294967295 blackhole
[...]
Fix by adjusting the dump callback to return zero when the dump is
complete. After the fix only one recvmsg() call is made and the
NLMSG_DONE message is appended to the RTM_NEWNEXTHOP response:
# ip nexthop add id $((2**32-1)) blackhole
# strace -e sendto,recvmsg -s 5 ip nexthop
sendto(3, [[{nlmsg_len=24, nlmsg_type=RTM_GETNEXTHOP, nlmsg_flags=NLM_F_REQUEST|NLM_F_DUMP, nlmsg_seq=1691394080, nlmsg_pid=0}, {nh_family=AF_UNSPEC, nh_scope=RT_SCOPE_UNIVERSE, nh_protocol=RTPROT_UNSPEC, nh_flags=0}], {nlmsg_len=0, nlmsg_type=0 /* NLMSG_??? */, nlmsg_flags=0, nlmsg_seq=0, nlmsg_pid=0}], 152, 0, NULL, 0) = 152
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[{iov_base=NULL, iov_len=0}], msg_iovlen=1, msg_controllen=0, msg_flags=MSG_TRUNC}, MSG_PEEK|MSG_TRUNC) = 56
recvmsg(3, {msg_name={sa_family=AF_NETLINK, nl_pid=0, nl_groups=00000000}, msg_namelen=12, msg_iov=[{iov_base=[[{nlmsg_len=36, nlmsg_type=RTM_NEWNEXTHOP, nlmsg_flags=NLM_F_MULTI, nlmsg_seq=1691394080, nlmsg_pid=342}, {nh_family=AF_INET, nh_scope=RT_SCOPE_UNIVERSE, nh_protocol=RTPROT_UNSPEC, nh_flags=0}, [[{nla_len=8, nla_type=NHA_ID}, 4294967295], {nla_len=4, nla_type=NHA_BLACKHOLE}]], [{nlmsg_len=20, nlmsg_type=NLMSG_DONE, nlmsg_flags=NLM_F_MULTI, nlmsg_seq=1691394080, nlmsg_pid=342}, 0]], iov_len=32768}], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 56
id 4294967295 blackhole
+++ exited with 0 +++
Note that if the NLMSG_DONE message cannot be appended because of size
limitations, then another recvmsg() will be needed, but the core netlink
code will not invoke the dump callback and simply reply with a
NLMSG_DONE message since it knows that the callback previously returned
zero.
Add a test that fails before the fix:
# ./fib_nexthops.sh -t basic
[...]
TEST: Maximum nexthop ID dump [FAIL]
[...]
And passes after it:
# ./fib_nexthops.sh -t basic
[...]
TEST: Maximum nexthop ID dump [ OK ]
[...]
Fixes: ab84be7e54 ("net: Initial nexthop code")
Reported-by: Petr Machata <petrm@nvidia.com>
Closes: https://lore.kernel.org/netdev/87sf91enuf.fsf@nvidia.com/
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20230808075233.3337922-2-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 6a7ac3d205 upstream.
If we try to emit an icmp error in response to a nonliner skb, we get
BUG: KASAN: slab-out-of-bounds in ip_compute_csum+0x134/0x220
Read of size 4 at addr ffff88811c50db00 by task iperf3/1691
CPU: 2 PID: 1691 Comm: iperf3 Not tainted 6.5.0-rc3+ #309
[..]
kasan_report+0x105/0x140
ip_compute_csum+0x134/0x220
iptunnel_pmtud_build_icmp+0x554/0x1020
skb_tunnel_check_pmtu+0x513/0xb80
vxlan_xmit_one+0x139e/0x2ef0
vxlan_xmit+0x1867/0x2760
dev_hard_start_xmit+0x1ee/0x4f0
br_dev_queue_push_xmit+0x4d1/0x660
[..]
ip_compute_csum() cannot deal with nonlinear skbs, so avoid it.
After this change, splat is gone and iperf3 is no longer stuck.
Fixes: 4cb47a8644 ("tunnels: PMTU discovery support for directly bridged IP packets")
Signed-off-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/r/20230803152653.29535-2-fw@strlen.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit ddf251fa2b ]
Whenever tcpm_new() reclaims an old entry, tcpm_suck_dst()
would overwrite data that could be read from tcp_fastopen_cache_get()
or tcp_metrics_fill_info().
We need to acquire fastopen_seqlock to maintain consistency.
For newly allocated objects, tcpm_new() can switch to kzalloc()
to avoid an extra fastopen_seqlock acquisition.
Fixes: 1fe4c481ba ("net-tcp: Fast Open client - cookie cache")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20230802131500.1478140-7-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit d5d986ce42 ]
tm->tcpm_net can be read or written locklessly.
Instead of changing write_pnet() and read_pnet() and potentially
hurt performance, add the needed READ_ONCE()/WRITE_ONCE()
in tm_net() and tcpm_new().
Fixes: 849e8a0ca8 ("tcp_metrics: Add a field tcpm_net and verify it matches on lookup")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20230802131500.1478140-6-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 8c4d04f6b4 ]
tm->tcpm_vals[] values can be read or written locklessly.
Add needed READ_ONCE()/WRITE_ONCE() to document this,
and force use of tcp_metric_get() and tcp_metric_set()
Fixes: 51c5d0c4b1 ("tcp: Maintain dynamic metrics in local cache.")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 285ce119a3 ]
tm->tcpm_lock can be read or written locklessly.
Add needed READ_ONCE()/WRITE_ONCE() to document this.
Fixes: 51c5d0c4b1 ("tcp: Maintain dynamic metrics in local cache.")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20230802131500.1478140-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 949ad62a5d ]
tm->tcpm_stamp can be read or written locklessly.
Add needed READ_ONCE()/WRITE_ONCE() to document this.
Also constify tcpm_check_stamp() dst argument.
Fixes: 51c5d0c4b1 ("tcp: Maintain dynamic metrics in local cache.")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20230802131500.1478140-3-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e6638094d7 ]
Because v4 and v6 families use separate inetpeer trees (respectively
net->ipv4.peers and net->ipv6.peers), inetpeer_addr_cmp(a, b) assumes
a & b share the same family.
tcp_metrics use a common hash table, where entries can have different
families.
We must therefore make sure to not call inetpeer_addr_cmp()
if the families do not match.
Fixes: d39d14ffa2 ("net: Add helper function to compare inetpeer addresses")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20230802131500.1478140-2-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 8bf43be799 ]
sk_getsockopt() runs locklessly. This means sk->sk_priority
can be read while other threads are changing its value.
Other reads also happen without socket lock being held.
Add missing annotations where needed.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 3c5b4d69c3 ]
sk->sk_mark is often read while another thread could change the value.
Fixes: 4a19ec5800 ("[NET]: Introducing socket mark socket option.")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 70f360dd70 ]
This field can be read locklessly.
Fixes: 1536e2857b ("tcp: Add a TCP_FASTOPEN socket option to get a max backlog on its listner")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230719212857.3943972-12-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 26023e91e1 ]
This field can be read locklessly from do_tcp_getsockopt()
Fixes: dca43c75e7 ("tcp: Add TCP_USER_TIMEOUT socket option.")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230719212857.3943972-11-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 3a037f0f3c ]
do_tcp_getsockopt() and reqsk_timer_handler() read
icsk->icsk_syn_retries while another cpu might change its value.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230719212857.3943972-7-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit dd23c9f1e8 ]
do_tcp_getsockopt() reads tp->tsoffset while another cpu
might change its value.
Fixes: 93be6ce0e9 ("tcp: set and get per-socket timestamp")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230719212857.3943972-3-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 81b3ade5d2 ]
This reverts commit 3f4ca5fafc.
Commit 3f4ca5fafc ("tcp: avoid the lookup process failing to get sk in
ehash table") reversed the order in how a socket is inserted into ehash
to fix an issue that ehash-lookup could fail when reqsk/full sk/twsk are
swapped. However, it introduced another lookup failure.
The full socket in ehash is allocated from a slab with SLAB_TYPESAFE_BY_RCU
and does not have SOCK_RCU_FREE, so the socket could be reused even while
it is being referenced on another CPU doing RCU lookup.
Let's say a socket is reused and inserted into the same hash bucket during
lookup. After the blamed commit, a new socket is inserted at the end of
the list. If that happens, we will skip sockets placed after the previous
position of the reused socket, resulting in ehash lookup failure.
As described in Documentation/RCU/rculist_nulls.rst, we should insert a
new socket at the head of the list to avoid such an issue.
This issue, the swap-lookup-failure, and another variant reported in [0]
can all be handled properly by adding a locked ehash lookup suggested by
Eric Dumazet [1].
However, this issue could occur for every packet, thus more likely than
the other two races, so let's revert the change for now.
Link: https://lore.kernel.org/netdev/20230606064306.9192-1-duanmuquan@baidu.com/ [0]
Link: https://lore.kernel.org/netdev/CANn89iK8snOz8TYOhhwfimC7ykYA78GA3Nyv8x06SZYa1nKdyA@mail.gmail.com/ [1]
Fixes: 3f4ca5fafc ("tcp: avoid the lookup process failing to get sk in ehash table")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20230717215918.15723-1-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit daa751444f ]
key might contain private part of the key, so better use
kfree_sensitive to free it.
Fixes: 38320c70d2 ("[IPSEC]: Use crypto_aead and authenc in ESP")
Signed-off-by: Wang Ming <machel@vivo.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit c0a8966e2b ]
When using IPv4/TCP, skb->hash comes from sk->sk_txhash except in
TIME_WAIT and SYN_RECV where it's not set in the reply skb from
ip_send_unicast_reply. Those packets will have a mismatched hash with
others from the same flow as their hashes will be 0. IPv6 does not have
the same issue as the hash is set from the socket txhash in those cases.
This commits sets the hash in the reply skb from ip_send_unicast_reply,
which makes the IPv4 code behaving like IPv6.
Signed-off-by: Antoine Tenart <atenart@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Stable-dep-of: 5e5265522a ("tcp: annotate data-races around tcp_rsk(req)->txhash")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 9378096e8a ]
This is a preparatory commit to replace `lock_sock_fast` with
`lock_sock`,and facilitate BPF programs executed from the TCP sockets
iterator to be able to destroy TCP sockets using the bpf_sock_destroy
kfunc (implemented in follow-up commits).
Previously, BPF TCP iterator was acquiring the sock lock with BH
disabled. This led to scenarios where the sockets hash table bucket lock
can be acquired with BH enabled in some path versus disabled in other.
In such situation, kernel issued a warning since it thinks that in the
BH enabled path the same bucket lock *might* be acquired again in the
softirq context (BH disabled), which will lead to a potential dead lock.
Since bpf_sock_destroy also happens in a process context, the potential
deadlock warning is likely a false alarm.
Here is a snippet of annotated stack trace that motivated this change:
```
Possible interrupt unsafe locking scenario:
CPU0 CPU1
---- ----
lock(&h->lhash2[i].lock);
local_bh_disable();
lock(&h->lhash2[i].lock);
kernel imagined possible scenario:
local_bh_disable(); /* Possible softirq */
lock(&h->lhash2[i].lock);
*** Potential Deadlock ***
process context:
lock_acquire+0xcd/0x330
_raw_spin_lock+0x33/0x40
------> Acquire (bucket) lhash2.lock with BH enabled
__inet_hash+0x4b/0x210
inet_csk_listen_start+0xe6/0x100
inet_listen+0x95/0x1d0
__sys_listen+0x69/0xb0
__x64_sys_listen+0x14/0x20
do_syscall_64+0x3c/0x90
entry_SYSCALL_64_after_hwframe+0x72/0xdc
bpf_sock_destroy run from iterator:
lock_acquire+0xcd/0x330
_raw_spin_lock+0x33/0x40
------> Acquire (bucket) lhash2.lock with BH disabled
inet_unhash+0x9a/0x110
tcp_set_state+0x6a/0x210
tcp_abort+0x10d/0x200
bpf_prog_6793c5ca50c43c0d_iter_tcp6_server+0xa4/0xa9
bpf_iter_run_prog+0x1ff/0x340
------> lock_sock_fast that acquires sock lock with BH disabled
bpf_iter_tcp_seq_show+0xca/0x190
bpf_seq_read+0x177/0x450
```
Also, Yonghong reported a deadlock for non-listening TCP sockets that
this change resolves. Previously, `lock_sock_fast` held the sock spin
lock with BH which was again being acquired in `tcp_abort`:
```
watchdog: BUG: soft lockup - CPU#0 stuck for 86s! [test_progs:2331]
RIP: 0010:queued_spin_lock_slowpath+0xd8/0x500
Call Trace:
<TASK>
_raw_spin_lock+0x84/0x90
tcp_abort+0x13c/0x1f0
bpf_prog_88539c5453a9dd47_iter_tcp6_client+0x82/0x89
bpf_iter_run_prog+0x1aa/0x2c0
? preempt_count_sub+0x1c/0xd0
? from_kuid_munged+0x1c8/0x210
bpf_iter_tcp_seq_show+0x14e/0x1b0
bpf_seq_read+0x36c/0x6a0
bpf_iter_tcp_seq_show
lock_sock_fast
__lock_sock_fast
spin_lock_bh(&sk->sk_lock.slock);
/* * Fast path return with bottom halves disabled and * sock::sk_lock.slock held.* */
...
tcp_abort
local_bh_disable();
spin_lock(&((sk)->sk_lock.slock)); // from bh_lock_sock(sk)
```
With the switch to `lock_sock`, it calls `spin_unlock_bh` before returning:
```
lock_sock
lock_sock_nested
spin_lock_bh(&sk->sk_lock.slock);
:
spin_unlock_bh(&sk->sk_lock.slock);
```
Acked-by: Yonghong Song <yhs@meta.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Aditi Ghag <aditi.ghag@isovalent.com>
Link: https://lore.kernel.org/r/20230519225157.760788-2-aditi.ghag@isovalent.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 998127cdb4 ]
request sockets are lockless, __tcp_oow_rate_limited() could be called
on the same object from different cpus. This is harmless.
Add READ_ONCE()/WRITE_ONCE() annotations to avoid a KCSAN report.
Fixes: 4ce7e93cb3 ("tcp: rate limit ACK sent by SYN_RECV request sockets")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit f015b900bc ]
With offloading enabled, esp_xmit() gets invoked very late, from within
validate_xmit_xfrm() which is after validate_xmit_skb() validates and
linearizes the skb if the underlying device does not support fragments.
esp_output_tail() may add a fragment to the skb while adding the auth
tag/ IV. Devices without the proper support will then send skb->data
points to with the correct length so the packet will have garbage at the
end. A pcap sniffer will claim that the proper data has been sent since
it parses the skb properly.
It is not affected with INET_ESP_OFFLOAD disabled.
Linearize the skb after offloading if the sending hardware requires it.
It was tested on v4, v6 has been adopted.
Fixes: 7785bba299 ("esp: Add a software GRO codepath")
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 1166a530a8 ]
Before Linux v5.8 an AF_INET6 SOCK_DGRAM (udp/udplite) socket
with SOL_UDP, UDP_ENCAP, UDP_ENCAP_ESPINUDP{,_NON_IKE} enabled
would just unconditionally use xfrm4_udp_encap_rcv(), afterwards
such a socket would use the newly added xfrm6_udp_encap_rcv()
which only handles IPv6 packets.
Cc: Sabrina Dubroca <sd@queasysnail.net>
Cc: Steffen Klassert <steffen.klassert@secunet.com>
Cc: Jakub Kicinski <kuba@kernel.org>
Cc: Benedict Wong <benedictwong@google.com>
Cc: Yan Yan <evitayan@google.com>
Fixes: 0146dca70b ("xfrm: add support for UDPv6 encapsulation of ESP")
Signed-off-by: Maciej Żenczykowski <maze@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 82a01ab35b ]
We missed that tcp_gso_segment() was assuming skb->len was smaller than 65535 :
oldlen = (u16)~skb->len;
This part came with commit 0718bcc09b ("[NET]: Fix CHECKSUM_HW GSO problems.")
This leads to wrong TCP checksum.
Adapt the code to accept arbitrary packet length.
v2:
- use two csum_add() instead of csum_fold() (Alexander Duyck)
- Change delta type to __wsum to reduce casts (Alexander Duyck)
Fixes: 09f3d1a3a5 ("ipv6/gso: remove temporary HBH/jumbo header")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Alexander Duyck <alexanderduyck@fb.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230605161647.3624428-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e209fee411 ]
With this commit, all the GIDs ("0 4294967294") can be written to the
"net.ipv4.ping_group_range" sysctl.
Note that 4294967295 (0xffffffff) is an invalid GID (see gid_valid() in
include/linux/uidgid.h), and an attempt to register this number will cause
-EINVAL.
Prior to this commit, only up to GID 2147483647 could be covered.
Documentation/networking/ip-sysctl.rst had "0 4294967295" as an example
value, but this example was wrong and causing -EINVAL.
Fixes: c319b4d76b ("net: ipv4: add IPPROTO_ICMP socket kind")
Co-developed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Akihiro Suda <akihiro.suda.cz@hco.ntt.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 30c6f0bf95 ]
In this patch, we mainly try to handle sending a compressed ack
correctly if it's deferred.
Here are more details in the old logic:
When sack compression is triggered in the tcp_compressed_ack_kick(),
if the sock is owned by user, it will set TCP_DELACK_TIMER_DEFERRED
and then defer to the release cb phrase. Later once user releases
the sock, tcp_delack_timer_handler() should send a ack as expected,
which, however, cannot happen due to lack of ICSK_ACK_TIMER flag.
Therefore, the receiver would not sent an ack until the sender's
retransmission timeout. It definitely increases unnecessary latency.
Fixes: 5d9f4262b7 ("tcp: add SACK compression")
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: fuyuanli <fuyuanli@didiglobal.com>
Signed-off-by: Jason Xing <kerneljasonxing@gmail.com>
Link: https://lore.kernel.org/netdev/20230529113804.GA20300@didi-ThinkCentre-M920t-N000/
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230531080150.GA20424@didi-ThinkCentre-M920t-N000
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e5c6de5fa0 ]
The read_skb() logic is incrementing the tcp->copied_seq which is used for
among other things calculating how many outstanding bytes can be read by
the application. This results in application errors, if the application
does an ioctl(FIONREAD) we return zero because this is calculated from
the copied_seq value.
To fix this we move tcp->copied_seq accounting into the recv handler so
that we update these when the recvmsg() hook is called and data is in
fact copied into user buffers. This gives an accurate FIONREAD value
as expected and improves ACK handling. Before we were calling the
tcp_rcv_space_adjust() which would update 'number of bytes copied to
user in last RTT' which is wrong for programs returning SK_PASS. The
bytes are only copied to the user when recvmsg is handled.
Doing the fix for recvmsg is straightforward, but fixing redirect and
SK_DROP pkts is a bit tricker. Build a tcp_psock_eat() helper and then
call this from skmsg handlers. This fixes another issue where a broken
socket with a BPF program doing a resubmit could hang the receiver. This
happened because although read_skb() consumed the skb through sock_drop()
it did not update the copied_seq. Now if a single reccv socket is
redirecting to many sockets (for example for lb) the receiver sk will be
hung even though we might expect it to continue. The hang comes from
not updating the copied_seq numbers and memory pressure resulting from
that.
We have a slight layer problem of calling tcp_eat_skb even if its not
a TCP socket. To fix we could refactor and create per type receiver
handlers. I decided this is more work than we want in the fix and we
already have some small tweaks depending on caller that use the
helper skb_bpf_strparser(). So we extend that a bit and always set
the strparser bit when it is in use and then we can gate the
seq_copied updates on this.
Fixes: 04919bed94 ("tcp: Introduce tcp_read_skb()")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Link: https://lore.kernel.org/bpf/20230523025618.113937-9-john.fastabend@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit ea444185a6 ]
A common mechanism to put a TCP socket into the sockmap is to hook the
BPF_SOCK_OPS_{ACTIVE_PASSIVE}_ESTABLISHED_CB event with a BPF program
that can map the socket info to the correct BPF verdict parser. When
the user adds the socket to the map the psock is created and the new
ops are assigned to ensure the verdict program will 'see' the sk_buffs
as they arrive.
Part of this process hooks the sk_data_ready op with a BPF specific
handler to wake up the BPF verdict program when data is ready to read.
The logic is simple enough (posted here for easy reading)
static void sk_psock_verdict_data_ready(struct sock *sk)
{
struct socket *sock = sk->sk_socket;
if (unlikely(!sock || !sock->ops || !sock->ops->read_skb))
return;
sock->ops->read_skb(sk, sk_psock_verdict_recv);
}
The oversight here is sk->sk_socket is not assigned until the application
accepts() the new socket. However, its entirely ok for the peer application
to do a connect() followed immediately by sends. The socket on the receiver
is sitting on the backlog queue of the listening socket until its accepted
and the data is queued up. If the peer never accepts the socket or is slow
it will eventually hit data limits and rate limit the session. But,
important for BPF sockmap hooks when this data is received TCP stack does
the sk_data_ready() call but the read_skb() for this data is never called
because sk_socket is missing. The data sits on the sk_receive_queue.
Then once the socket is accepted if we never receive more data from the
peer there will be no further sk_data_ready calls and all the data
is still on the sk_receive_queue(). Then user calls recvmsg after accept()
and for TCP sockets in sockmap we use the tcp_bpf_recvmsg_parser() handler.
The handler checks for data in the sk_msg ingress queue expecting that
the BPF program has already run from the sk_data_ready hook and enqueued
the data as needed. So we are stuck.
To fix do an unlikely check in recvmsg handler for data on the
sk_receive_queue and if it exists wake up data_ready. We have the sock
locked in both read_skb and recvmsg so should avoid having multiple
runners.
Fixes: 04919bed94 ("tcp: Introduce tcp_read_skb()")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Link: https://lore.kernel.org/bpf/20230523025618.113937-7-john.fastabend@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 901546fd8f ]
The sockmap code is returning EAGAIN after a FIN packet is received and no
more data is on the receive queue. Correct behavior is to return 0 to the
user and the user can then close the socket. The EAGAIN causes many apps
to retry which masks the problem. Eventually the socket is evicted from
the sockmap because its released from sockmap sock free handling. The
issue creates a delay and can cause some errors on application side.
To fix this check on sk_msg_recvmsg side if length is zero and FIN flag
is set then set return to zero. A selftest will be added to check this
condition.
Fixes: 04919bed94 ("tcp: Introduce tcp_read_skb()")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: William Findlay <will@isovalent.com>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Link: https://lore.kernel.org/bpf/20230523025618.113937-6-john.fastabend@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 78fa0d61d9 ]
The read_skb hook calls consume_skb() now, but this means that if the
recv_actor program wants to use the skb it needs to inc the ref cnt
so that the consume_skb() doesn't kfree the sk_buff.
This is problematic because in some error cases under memory pressure
we may need to linearize the sk_buff from sk_psock_skb_ingress_enqueue().
Then we get this,
skb_linearize()
__pskb_pull_tail()
pskb_expand_head()
BUG_ON(skb_shared(skb))
Because we incremented users refcnt from sk_psock_verdict_recv() we
hit the bug on with refcnt > 1 and trip it.
To fix lets simply pass ownership of the sk_buff through the skb_read
call. Then we can drop the consume from read_skb handlers and assume
the verdict recv does any required kfree.
Bug found while testing in our CI which runs in VMs that hit memory
constraints rather regularly. William tested TCP read_skb handlers.
[ 106.536188] ------------[ cut here ]------------
[ 106.536197] kernel BUG at net/core/skbuff.c:1693!
[ 106.536479] invalid opcode: 0000 [#1] PREEMPT SMP PTI
[ 106.536726] CPU: 3 PID: 1495 Comm: curl Not tainted 5.19.0-rc5 #1
[ 106.537023] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS ArchLinux 1.16.0-1 04/01/2014
[ 106.537467] RIP: 0010:pskb_expand_head+0x269/0x330
[ 106.538585] RSP: 0018:ffffc90000138b68 EFLAGS: 00010202
[ 106.538839] RAX: 000000000000003f RBX: ffff8881048940e8 RCX: 0000000000000a20
[ 106.539186] RDX: 0000000000000002 RSI: 0000000000000000 RDI: ffff8881048940e8
[ 106.539529] RBP: ffffc90000138be8 R08: 00000000e161fd1a R09: 0000000000000000
[ 106.539877] R10: 0000000000000018 R11: 0000000000000000 R12: ffff8881048940e8
[ 106.540222] R13: 0000000000000003 R14: 0000000000000000 R15: ffff8881048940e8
[ 106.540568] FS: 00007f277dde9f00(0000) GS:ffff88813bd80000(0000) knlGS:0000000000000000
[ 106.540954] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 106.541227] CR2: 00007f277eeede64 CR3: 000000000ad3e000 CR4: 00000000000006e0
[ 106.541569] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 106.541915] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 106.542255] Call Trace:
[ 106.542383] <IRQ>
[ 106.542487] __pskb_pull_tail+0x4b/0x3e0
[ 106.542681] skb_ensure_writable+0x85/0xa0
[ 106.542882] sk_skb_pull_data+0x18/0x20
[ 106.543084] bpf_prog_b517a65a242018b0_bpf_skskb_http_verdict+0x3a9/0x4aa9
[ 106.543536] ? migrate_disable+0x66/0x80
[ 106.543871] sk_psock_verdict_recv+0xe2/0x310
[ 106.544258] ? sk_psock_write_space+0x1f0/0x1f0
[ 106.544561] tcp_read_skb+0x7b/0x120
[ 106.544740] tcp_data_queue+0x904/0xee0
[ 106.544931] tcp_rcv_established+0x212/0x7c0
[ 106.545142] tcp_v4_do_rcv+0x174/0x2a0
[ 106.545326] tcp_v4_rcv+0xe70/0xf60
[ 106.545500] ip_protocol_deliver_rcu+0x48/0x290
[ 106.545744] ip_local_deliver_finish+0xa7/0x150
Fixes: 04919bed94 ("tcp: Introduce tcp_read_skb()")
Reported-by: William Findlay <will@isovalent.com>
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: William Findlay <will@isovalent.com>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Link: https://lore.kernel.org/bpf/20230523025618.113937-2-john.fastabend@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 3632679d9e ]
With a raw socket bound to IPPROTO_RAW (ie with hdrincl enabled), the
protocol field of the flow structure, build by raw_sendmsg() /
rawv6_sendmsg()), is set to IPPROTO_RAW. This breaks the ipsec policy
lookup when some policies are defined with a protocol in the selector.
For ipv6, the sin6_port field from 'struct sockaddr_in6' could be used to
specify the protocol. Just accept all values for IPPROTO_RAW socket.
For ipv4, the sin_port field of 'struct sockaddr_in' could not be used
without breaking backward compatibility (the value of this field was never
checked). Let's add a new kind of control message, so that the userland
could specify which protocol is used.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
CC: stable@vger.kernel.org
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Link: https://lore.kernel.org/r/20230522120820.1319391-1-nicolas.dichtel@6wind.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 91d0b78c51 ]
Users who want to share a single public IP address for outgoing connections
between several hosts traditionally reach for SNAT. However, SNAT requires
state keeping on the node(s) performing the NAT.
A stateless alternative exists, where a single IP address used for egress
can be shared between several hosts by partitioning the available ephemeral
port range. In such a setup:
1. Each host gets assigned a disjoint range of ephemeral ports.
2. Applications open connections from the host-assigned port range.
3. Return traffic gets routed to the host based on both, the destination IP
and the destination port.
An application which wants to open an outgoing connection (connect) from a
given port range today can choose between two solutions:
1. Manually pick the source port by bind()'ing to it before connect()'ing
the socket.
This approach has a couple of downsides:
a) Search for a free port has to be implemented in the user-space. If
the chosen 4-tuple happens to be busy, the application needs to retry
from a different local port number.
Detecting if 4-tuple is busy can be either easy (TCP) or hard
(UDP). In TCP case, the application simply has to check if connect()
returned an error (EADDRNOTAVAIL). That is assuming that the local
port sharing was enabled (REUSEADDR) by all the sockets.
# Assume desired local port range is 60_000-60_511
s = socket(AF_INET, SOCK_STREAM)
s.setsockopt(SOL_SOCKET, SO_REUSEADDR, 1)
s.bind(("192.0.2.1", 60_000))
s.connect(("1.1.1.1", 53))
# Fails only if 192.0.2.1:60000 -> 1.1.1.1:53 is busy
# Application must retry with another local port
In case of UDP, the network stack allows binding more than one socket
to the same 4-tuple, when local port sharing is enabled
(REUSEADDR). Hence detecting the conflict is much harder and involves
querying sock_diag and toggling the REUSEADDR flag [1].
b) For TCP, bind()-ing to a port within the ephemeral port range means
that no connecting sockets, that is those which leave it to the
network stack to find a free local port at connect() time, can use
the this port.
IOW, the bind hash bucket tb->fastreuse will be 0 or 1, and the port
will be skipped during the free port search at connect() time.
2. Isolate the app in a dedicated netns and use the use the per-netns
ip_local_port_range sysctl to adjust the ephemeral port range bounds.
The per-netns setting affects all sockets, so this approach can be used
only if:
- there is just one egress IP address, or
- the desired egress port range is the same for all egress IP addresses
used by the application.
For TCP, this approach avoids the downsides of (1). Free port search and
4-tuple conflict detection is done by the network stack:
system("sysctl -w net.ipv4.ip_local_port_range='60000 60511'")
s = socket(AF_INET, SOCK_STREAM)
s.setsockopt(SOL_IP, IP_BIND_ADDRESS_NO_PORT, 1)
s.bind(("192.0.2.1", 0))
s.connect(("1.1.1.1", 53))
# Fails if all 4-tuples 192.0.2.1:60000-60511 -> 1.1.1.1:53 are busy
For UDP this approach has limited applicability. Setting the
IP_BIND_ADDRESS_NO_PORT socket option does not result in local source
port being shared with other connected UDP sockets.
Hence relying on the network stack to find a free source port, limits the
number of outgoing UDP flows from a single IP address down to the number
of available ephemeral ports.
To put it another way, partitioning the ephemeral port range between hosts
using the existing Linux networking API is cumbersome.
To address this use case, add a new socket option at the SOL_IP level,
named IP_LOCAL_PORT_RANGE. The new option can be used to clamp down the
ephemeral port range for each socket individually.
The option can be used only to narrow down the per-netns local port
range. If the per-socket range lies outside of the per-netns range, the
latter takes precedence.
UAPI-wise, the low and high range bounds are passed to the kernel as a pair
of u16 values in host byte order packed into a u32. This avoids pointer
passing.
PORT_LO = 40_000
PORT_HI = 40_511
s = socket(AF_INET, SOCK_STREAM)
v = struct.pack("I", PORT_HI << 16 | PORT_LO)
s.setsockopt(SOL_IP, IP_LOCAL_PORT_RANGE, v)
s.bind(("127.0.0.1", 0))
s.getsockname()
# Local address between ("127.0.0.1", 40_000) and ("127.0.0.1", 40_511),
# if there is a free port. EADDRINUSE otherwise.
[1] https://github.com/cloudflare/cloudflare-blog/blob/232b432c1d57/2022-02-connectx/connectx.py#L116
Reviewed-by: Marek Majkowski <marek@cloudflare.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Stable-dep-of: 3632679d9e ("ipv{4,6}/raw: fix output xfrm lookup wrt protocol")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 1e306ec49a ]
When tcp_v4_send_reset() is called with @sk == NULL,
we do not change ctl_sk->sk_priority, which could have been
set from a prior invocation.
Change tcp_v4_send_reset() to set sk_priority and sk_mark
fields before calling ip_send_unicast_reply().
This means tcp_v4_send_reset() and tcp_v4_send_ack()
no longer have to clear ctl_sk->sk_mark after
their call to ip_send_unicast_reply().
Fixes: f6c0f5d209 ("tcp: honor SO_PRIORITY in TIME_WAIT state")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Antoine Tenart <atenart@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 99e5acae19 ]
Like commit ea30388bae ("ipv6: Fix an uninit variable access bug in
__ip6_make_skb()"). icmphdr does not in skb linear region under the
scenario of SOCK_RAW socket. Access icmp_hdr(skb)->type directly will
trigger the uninit variable access bug.
Use a local variable icmp_type to carry the correct value in different
scenarios.
Fixes: 96793b4825 ("[IPV4]: Add ICMPMsgStats MIB (RFC 4293)")
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Ziyang Xuan <william.xuanziyang@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit dc5110c2d9 ]
UBSAN: shift-out-of-bounds in net/ipv4/tcp_input.c:555:23
shift exponent 255 is too large for 32-bit type 'int'
CPU: 1 PID: 7907 Comm: ssh Not tainted 6.3.0-rc4-00161-g62bad54b26db-dirty #206
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
Call Trace:
<TASK>
dump_stack_lvl+0x136/0x150
__ubsan_handle_shift_out_of_bounds+0x21f/0x5a0
tcp_init_transfer.cold+0x3a/0xb9
tcp_finish_connect+0x1d0/0x620
tcp_rcv_state_process+0xd78/0x4d60
tcp_v4_do_rcv+0x33d/0x9d0
__release_sock+0x133/0x3b0
release_sock+0x58/0x1b0
'maxwin' is int, shifting int for 32 or more bits is undefined behaviour.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 580031ff99 ]
While reviewing the udp-iter batching patches, noticed the bpf_iter_tcp
calling sock_put() is incorrect. It should call sock_gen_put instead
because bpf_iter_tcp is iterating the ehash table which has the req sk
and tw sk. This patch replaces all sock_put with sock_gen_put in the
bpf_iter_tcp codepath.
Fixes: 04c7820b77 ("bpf: tcp: Bpf iter batching and lock_sock")
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230328004232.2134233-1-martin.lau@linux.dev
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit ab5fb73ffa ]
After commit dbca1596bb ("ping: convert to RCU lookups, get rid
of rwlock"), we use RCU for ping sockets, but we should use spinlock
for /proc/net/icmp to avoid a potential NULL deref mentioned in
the previous patch.
Let's go back to using spinlock there.
Note we can convert ping sockets to use hlist instead of hlist_nulls
because we do not use SLAB_TYPESAFE_BY_RCU for ping sockets.
Fixes: dbca1596bb ("ping: convert to RCU lookups, get rid of rwlock")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 6579f5bacc ]
Some applications seem to rely on RAW sockets.
If they use private netns, we can avoid piling all RAW
sockets bound to a given protocol into a single bucket.
Also place (struct raw_hashinfo).lock into its own
cache line to limit false sharing.
Alternative would be to have per-netns hashtables,
but this seems too expensive for most netns
where RAW sockets are not used.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Stable-dep-of: 0a78cf7264 ("raw: Fix NULL deref in raw_get_next().")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 8a2618e14f ]
Commit f96a3d7455 ("ipv4: Fix incorrect route flushing when source
address is deleted") started to take the table ID field in the FIB info
structure into account when determining if two structures are identical
or not. This field is initialized using the 'fc_table' field in the
route configuration structure, which is not set when adding a route via
IOCTL.
The above can result in user space being able to install two identical
routes that only differ in the table ID field of their associated FIB
info.
Fix by initializing the table ID field in the route configuration
structure in the IOCTL path.
Before the fix:
# ip route add default via 192.0.2.2
# route add default gw 192.0.2.2
# ip -4 r show default
# default via 192.0.2.2 dev dummy10
# default via 192.0.2.2 dev dummy10
After the fix:
# ip route add default via 192.0.2.2
# route add default gw 192.0.2.2
SIOCADDRT: File exists
# ip -4 r show default
default via 192.0.2.2 dev dummy10
Audited the code paths to ensure there are no other paths that do not
properly initialize the route configuration structure when installing a
route.
Fixes: 5a56a0b3a4 ("net: Don't delete routes in different VRFs")
Fixes: f96a3d7455 ("ipv4: Fix incorrect route flushing when source address is deleted")
Reported-by: gaoxingwang <gaoxingwang1@huawei.com>
Link: https://lore.kernel.org/netdev/20230314144159.2354729-1-gaoxingwang1@huawei.com/
Tested-by: gaoxingwang <gaoxingwang1@huawei.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20230315124009.4015212-1-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit d9ba993428 ]
Paul Holzinger reported [0] that commit 5456262d2b ("net: Fix
incorrect address comparison when searching for a bind2 bucket")
introduced a bind() regression. Paul also gave a nice repro that
calls two types of bind() on the same port, both of which now
succeed, but the second call should fail:
bind(fd1, ::, port) + bind(fd2, 127.0.0.1, port)
The cited commit added address family tests in three functions to
fix the uninit-value KMSAN report. [1] However, the test added to
inet_bind2_bucket_match_addr_any() removed a necessary conflict
check; the dual-stack wildcard address no longer conflicts with
an IPv4 non-wildcard address.
If tb->family is AF_INET6 and sk->sk_family is AF_INET in
inet_bind2_bucket_match_addr_any(), we still need to check
if tb has the dual-stack wildcard address.
Note that the IPv4 wildcard address does not conflict with
IPv6 non-wildcard addresses.
[0]: https://lore.kernel.org/netdev/e21bf153-80b0-9ec0-15ba-e04a4ad42c34@redhat.com/
[1]: https://lore.kernel.org/netdev/CAG_fn=Ud3zSW7AZWXc+asfMhZVL5ETnvuY44Pmyv4NPv-ijN-A@mail.gmail.com/
Fixes: 5456262d2b ("net: Fix incorrect address comparison when searching for a bind2 bucket")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reported-by: Paul Holzinger <pholzing@redhat.com>
Link: https://lore.kernel.org/netdev/CAG_fn=Ud3zSW7AZWXc+asfMhZVL5ETnvuY44Pmyv4NPv-ijN-A@mail.gmail.com/
Reviewed-by: Eric Dumazet <edumazet@google.com>
Tested-by: Paul Holzinger <pholzing@redhat.com>
Reviewed-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
