rel-38
506 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 Paolo Abeni
|
497c1b7391 |
mptcp: prevent excessive coalescing on receive
BugLink: https://bugs.launchpad.net/bugs/2114239
commit 56b824eb49d6258aa0bad09a406ceac3f643cdae upstream.
Currently the skb size after coalescing is only limited by the skb
layout (the skb must not carry frag_list). A single coalesced skb
covering several MSS can potentially fill completely the receive
buffer. In such a case, the snd win will zero until the receive buffer
will be empty again, affecting tput badly.
Fixes:
|
||
|
Paolo Abeni
|
aea14d7cd2 |
mptcp: handle fastopen disconnect correctly
BugLink: https://bugs.launchpad.net/bugs/2111953
commit 619af16b3b57a3a4ee50b9a30add9ff155541e71 upstream.
Syzbot was able to trigger a data stream corruption:
WARNING: CPU: 0 PID: 9846 at net/mptcp/protocol.c:1024 __mptcp_clean_una+0xddb/0xff0 net/mptcp/protocol.c:1024
Modules linked in:
CPU: 0 UID: 0 PID: 9846 Comm: syz-executor351 Not tainted 6.13.0-rc2-syzkaller-00059-g00a5acdbf398 #0
Hardware name: Google Compute Engine/Google Compute Engine, BIOS Google 11/25/2024
RIP: 0010:__mptcp_clean_una+0xddb/0xff0 net/mptcp/protocol.c:1024
Code: fa ff ff 48 8b 4c 24 18 80 e1 07 fe c1 38 c1 0f 8c 8e fa ff ff 48 8b 7c 24 18 e8 e0 db 54 f6 e9 7f fa ff ff e8 e6 80 ee f5 90 <0f> 0b 90 4c 8b 6c 24 40 4d 89 f4 e9 04 f5 ff ff 44 89 f1 80 e1 07
RSP: 0018:ffffc9000c0cf400 EFLAGS: 00010293
RAX: ffffffff8bb0dd5a RBX: ffff888033f5d230 RCX: ffff888059ce8000
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
RBP: ffffc9000c0cf518 R08: ffffffff8bb0d1dd R09: 1ffff110170c8928
R10: dffffc0000000000 R11: ffffed10170c8929 R12: 0000000000000000
R13: ffff888033f5d220 R14: dffffc0000000000 R15: ffff8880592b8000
FS: 00007f6e866496c0(0000) GS:ffff8880b8600000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f6e86f491a0 CR3: 00000000310e6000 CR4: 00000000003526f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
__mptcp_clean_una_wakeup+0x7f/0x2d0 net/mptcp/protocol.c:1074
mptcp_release_cb+0x7cb/0xb30 net/mptcp/protocol.c:3493
release_sock+0x1aa/0x1f0 net/core/sock.c:3640
inet_wait_for_connect net/ipv4/af_inet.c:609 [inline]
__inet_stream_connect+0x8bd/0xf30 net/ipv4/af_inet.c:703
mptcp_sendmsg_fastopen+0x2a2/0x530 net/mptcp/protocol.c:1755
mptcp_sendmsg+0x1884/0x1b10 net/mptcp/protocol.c:1830
sock_sendmsg_nosec net/socket.c:711 [inline]
__sock_sendmsg+0x1a6/0x270 net/socket.c:726
____sys_sendmsg+0x52a/0x7e0 net/socket.c:2583
___sys_sendmsg net/socket.c:2637 [inline]
__sys_sendmsg+0x269/0x350 net/socket.c:2669
do_syscall_x64 arch/x86/entry/common.c:52 [inline]
do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83
entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f6e86ebfe69
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 b1 1f 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 b0 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f6e86649168 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
RAX: ffffffffffffffda RBX: 00007f6e86f491b8 RCX: 00007f6e86ebfe69
RDX: 0000000030004001 RSI: 0000000020000080 RDI: 0000000000000003
RBP: 00007f6e86f491b0 R08: 00007f6e866496c0 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00007f6e86f491bc
R13: 000000000000006e R14: 00007ffe445d9420 R15: 00007ffe445d9508
</TASK>
The root cause is the bad handling of disconnect() generated internally
by the MPTCP protocol in case of connect FASTOPEN errors.
Address the issue increasing the socket disconnect counter even on such
a case, to allow other threads waiting on the same socket lock to
properly error out.
Fixes:
|
||
|
Paolo Abeni
|
c3ed67f8da |
mptcp: don't always assume copied data in mptcp_cleanup_rbuf()
BugLink: https://bugs.launchpad.net/bugs/2106632
commit 551844f26da2a9f76c0a698baaffa631d1178645 upstream.
Under some corner cases the MPTCP protocol can end-up invoking
mptcp_cleanup_rbuf() when no data has been copied, but such helper
assumes the opposite condition.
Explicitly drop such assumption and performs the costly call only
when strictly needed - before releasing the msk socket lock.
Fixes:
|
||
|
Paolo Abeni
|
1ba715f1ed |
mptcp: fix recvbuffer adjust on sleeping rcvmsg
BugLink: https://bugs.launchpad.net/bugs/2106632 commit 449e6912a2522af672e99992e1201a454910864e upstream. If the recvmsg() blocks after receiving some data - i.e. due to SO_RCVLOWAT - the MPTCP code will attempt multiple times to adjust the receive buffer size, wrongly accounting every time the cumulative of received data - instead of accounting only for the delta. Address the issue moving mptcp_rcv_space_adjust just after the data reception and passing it only the just received bytes. This also removes an unneeded difference between the TCP and MPTCP RX code path implementation. Fixes: 581302298524 ("mptcp: error out earlier on disconnect") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni <pabeni@redhat.com> Reviewed-by: Mat Martineau <martineau@kernel.org> Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org> Link: https://patch.msgid.link/20241230-net-mptcp-rbuf-fixes-v1-1-8608af434ceb@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Manuel Diewald <manuel.diewald@canonical.com> Signed-off-by: Mehmet Basaran <mehmet.basaran@canonical.com> |
||
 Gang Yan
|
97c0ee1614 |
mptcp: annotate data-races around subflow->fully_established
BugLink: https://bugs.launchpad.net/bugs/2102118 [ Upstream commit 581c8cbfa934aaa555daa4e843242fcecc160f05 ] We introduce the same handling for potential data races with the 'fully_established' flag in subflow as previously done for msk->fully_established. Additionally, we make a crucial change: convert the subflow's 'fully_established' from 'bit_field' to 'bool' type. This is necessary because methods for avoiding data races don't work well with 'bit_field'. Specifically, the 'READ_ONCE' needs to know the size of the variable being accessed, which is not supported in 'bit_field'. Also, 'test_bit' expect the address of 'bit_field'. Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/516 Signed-off-by: Gang Yan <yangang@kylinos.cn> Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org> Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org> Link: https://patch.msgid.link/20241021-net-next-mptcp-misc-6-13-v1-2-1ef02746504a@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org> [koichiroden: adjusted mptcp_subflow_context bit field due to missing commit: 27069e7cb3d1 ("mptcp: disable active MPTCP in case of blackhole")] Signed-off-by: Koichiro Den <koichiro.den@canonical.com> Signed-off-by: Stefan Bader <stefan.bader@canonical.com> |
||
 Dmitry Kandybka
|
d818e97486 |
mptcp: fix possible integer overflow in mptcp_reset_tout_timer
BugLink: https://bugs.launchpad.net/bugs/2101915
commit b169e76ebad22cbd055101ee5aa1a7bed0e66606 upstream.
In 'mptcp_reset_tout_timer', promote 'probe_timestamp' to unsigned long
to avoid possible integer overflow. Compile tested only.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Signed-off-by: Dmitry Kandybka <d.kandybka@gmail.com>
Link: https://patch.msgid.link/20241107103657.1560536-1-d.kandybka@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
[ Conflict in this version because commit
|
||
|
Paolo Abeni
|
47b06d77f7 |
mptcp: cope racing subflow creation in mptcp_rcv_space_adjust
BugLink: https://bugs.launchpad.net/bugs/2101042
[ Upstream commit ce7356ae35943cc6494cc692e62d51a734062b7d ]
Additional active subflows - i.e. created by the in kernel path
manager - are included into the subflow list before starting the
3whs.
A racing recvmsg() spooling data received on an already established
subflow would unconditionally call tcp_cleanup_rbuf() on all the
current subflows, potentially hitting a divide by zero error on
the newly created ones.
Explicitly check that the subflow is in a suitable state before
invoking tcp_cleanup_rbuf().
Fixes:
|
||
|
Paolo Abeni
|
d5b260efeb |
mptcp: error out earlier on disconnect
BugLink: https://bugs.launchpad.net/bugs/2101042
[ Upstream commit 581302298524e9d77c4c44ff5156a6cd112227ae ]
Eric reported a division by zero splat in the MPTCP protocol:
Oops: divide error: 0000 [#1] PREEMPT SMP KASAN PTI
CPU: 1 UID: 0 PID: 6094 Comm: syz-executor317 Not tainted
6.12.0-rc5-syzkaller-00291-g05b92660cdfe #0
Hardware name: Google Google Compute Engine/Google Compute Engine,
BIOS Google 09/13/2024
RIP: 0010:__tcp_select_window+0x5b4/0x1310 net/ipv4/tcp_output.c:3163
Code: f6 44 01 e3 89 df e8 9b 75 09 f8 44 39 f3 0f 8d 11 ff ff ff e8
0d 74 09 f8 45 89 f4 e9 04 ff ff ff e8 00 74 09 f8 44 89 f0 99 <f7> 7c
24 14 41 29 d6 45 89 f4 e9 ec fe ff ff e8 e8 73 09 f8 48 89
RSP: 0018:ffffc900041f7930 EFLAGS: 00010293
RAX: 0000000000017e67 RBX: 0000000000017e67 RCX: ffffffff8983314b
RDX: 0000000000000000 RSI: ffffffff898331b0 RDI: 0000000000000004
RBP: 00000000005d6000 R08: 0000000000000004 R09: 0000000000017e67
R10: 0000000000003e80 R11: 0000000000000000 R12: 0000000000003e80
R13: ffff888031d9b440 R14: 0000000000017e67 R15: 00000000002eb000
FS: 00007feb5d7f16c0(0000) GS:ffff8880b8700000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007feb5d8adbb8 CR3: 0000000074e4c000 CR4: 00000000003526f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
__tcp_cleanup_rbuf+0x3e7/0x4b0 net/ipv4/tcp.c:1493
mptcp_rcv_space_adjust net/mptcp/protocol.c:2085 [inline]
mptcp_recvmsg+0x2156/0x2600 net/mptcp/protocol.c:2289
inet_recvmsg+0x469/0x6a0 net/ipv4/af_inet.c:885
sock_recvmsg_nosec net/socket.c:1051 [inline]
sock_recvmsg+0x1b2/0x250 net/socket.c:1073
__sys_recvfrom+0x1a5/0x2e0 net/socket.c:2265
__do_sys_recvfrom net/socket.c:2283 [inline]
__se_sys_recvfrom net/socket.c:2279 [inline]
__x64_sys_recvfrom+0xe0/0x1c0 net/socket.c:2279
do_syscall_x64 arch/x86/entry/common.c:52 [inline]
do_syscall_64+0xcd/0x250 arch/x86/entry/common.c:83
entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7feb5d857559
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 51 18 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 b0 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007feb5d7f1208 EFLAGS: 00000246 ORIG_RAX: 000000000000002d
RAX: ffffffffffffffda RBX: 00007feb5d8e1318 RCX: 00007feb5d857559
RDX: 000000800000000e RSI: 0000000000000000 RDI: 0000000000000003
RBP: 00007feb5d8e1310 R08: 0000000000000000 R09: ffffffff81000000
R10: 0000000000000100 R11: 0000000000000246 R12: 00007feb5d8e131c
R13: 00007feb5d8ae074 R14: 000000800000000e R15: 00000000fffffdef
and provided a nice reproducer.
The root cause is the current bad handling of racing disconnect.
After the blamed commit below, sk_wait_data() can return (with
error) with the underlying socket disconnected and a zero rcv_mss.
Catch the error and return without performing any additional
operations on the current socket.
Reported-by: Eric Dumazet <edumazet@google.com>
Fixes:
|
||
 Matthieu Baerts (NGI0)
|
6e96d105c1 |
mptcp: init: protect sched with rcu_read_lock
BugLink: https://bugs.launchpad.net/bugs/2099996
[ Upstream commit 3deb12c788c385e17142ce6ec50f769852fcec65 ]
Enabling CONFIG_PROVE_RCU_LIST with its dependence CONFIG_RCU_EXPERT
creates this splat when an MPTCP socket is created:
=============================
WARNING: suspicious RCU usage
6.12.0-rc2+ #11 Not tainted
-----------------------------
net/mptcp/sched.c:44 RCU-list traversed in non-reader section!!
other info that might help us debug this:
rcu_scheduler_active = 2, debug_locks = 1
no locks held by mptcp_connect/176.
stack backtrace:
CPU: 0 UID: 0 PID: 176 Comm: mptcp_connect Not tainted 6.12.0-rc2+ #11
Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
Call Trace:
<TASK>
dump_stack_lvl (lib/dump_stack.c:123)
lockdep_rcu_suspicious (kernel/locking/lockdep.c:6822)
mptcp_sched_find (net/mptcp/sched.c:44 (discriminator 7))
mptcp_init_sock (net/mptcp/protocol.c:2867 (discriminator 1))
? sock_init_data_uid (arch/x86/include/asm/atomic.h:28)
inet_create.part.0.constprop.0 (net/ipv4/af_inet.c:386)
? __sock_create (include/linux/rcupdate.h:347 (discriminator 1))
__sock_create (net/socket.c:1576)
__sys_socket (net/socket.c:1671)
? __pfx___sys_socket (net/socket.c:1712)
? do_user_addr_fault (arch/x86/mm/fault.c:1419 (discriminator 1))
__x64_sys_socket (net/socket.c:1728)
do_syscall_64 (arch/x86/entry/common.c:52 (discriminator 1))
entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)
That's because when the socket is initialised, rcu_read_lock() is not
used despite the explicit comment written above the declaration of
mptcp_sched_find() in sched.c. Adding the missing lock/unlock avoids the
warning.
Fixes:
|
||
|
Paolo Abeni
|
47fa54700e |
mptcp: handle consistently DSS corruption
BugLink: https://bugs.launchpad.net/bugs/2097301
commit e32d262c89e2b22cb0640223f953b548617ed8a6 upstream.
Bugged peer implementation can send corrupted DSS options, consistently
hitting a few warning in the data path. Use DEBUG_NET assertions, to
avoid the splat on some builds and handle consistently the error, dumping
related MIBs and performing fallback and/or reset according to the
subflow type.
Fixes:
|
||
|
Matthieu Baerts (NGI0)
|
206df36df9 |
mptcp: avoid duplicated SUB_CLOSED events
BugLink: https://bugs.launchpad.net/bugs/2084225
commit d82809b6c5f2676b382f77a5cbeb1a5d91ed2235 upstream.
The initial subflow might have already been closed, but still in the
connection list. When the worker is instructed to close the subflows
that have been marked as closed, it might then try to close the initial
subflow again.
A consequence of that is that the SUB_CLOSED event can be seen twice:
# ip mptcp endpoint
1.1.1.1 id 1 subflow dev eth0
2.2.2.2 id 2 subflow dev eth1
# ip mptcp monitor &
[ CREATED] remid=0 locid=0 saddr4=1.1.1.1 daddr4=9.9.9.9
[ ESTABLISHED] remid=0 locid=0 saddr4=1.1.1.1 daddr4=9.9.9.9
[ SF_ESTABLISHED] remid=0 locid=2 saddr4=2.2.2.2 daddr4=9.9.9.9
# ip mptcp endpoint delete id 1
[ SF_CLOSED] remid=0 locid=0 saddr4=1.1.1.1 daddr4=9.9.9.9
[ SF_CLOSED] remid=0 locid=0 saddr4=1.1.1.1 daddr4=9.9.9.9
The first one is coming from mptcp_pm_nl_rm_subflow_received(), and the
second one from __mptcp_close_subflow().
To avoid doing the post-closed processing twice, the subflow is now
marked as closed the first time.
Note that it is not enough to check if we are dealing with the first
subflow and check its sk_state: the subflow might have been reset or
closed before calling mptcp_close_ssk().
Fixes:
|
||
|
Matthieu Baerts (NGI0)
|
944f7d306e |
mptcp: pr_debug: add missing \n at the end
BugLink: https://bugs.launchpad.net/bugs/2084225
commit cb41b195e634d3f1ecfcd845314e64fd4bb3c7aa upstream.
pr_debug() have been added in various places in MPTCP code to help
developers to debug some situations. With the dynamic debug feature, it
is easy to enable all or some of them, and asks users to reproduce
issues with extra debug.
Many of these pr_debug() don't end with a new line, while no 'pr_cont()'
are used in MPTCP code. So the goal was not to display multiple debug
messages on one line: they were then not missing the '\n' on purpose.
Not having the new line at the end causes these messages to be printed
with a delay, when something else needs to be printed. This issue is not
visible when many messages need to be printed, but it is annoying and
confusing when only specific messages are expected, e.g.
# echo "func mptcp_pm_add_addr_echoed +fmp" \
> /sys/kernel/debug/dynamic_debug/control
# ./mptcp_join.sh "signal address"; \
echo "$(awk '{print $1}' /proc/uptime) - end"; \
sleep 5s; \
echo "$(awk '{print $1}' /proc/uptime) - restart"; \
./mptcp_join.sh "signal address"
013 signal address
(...)
10.75 - end
15.76 - restart
013 signal address
[ 10.367935] mptcp:mptcp_pm_add_addr_echoed: MPTCP: msk=(...)
(...)
=> a delay of 5 seconds: printed with a 10.36 ts, but after 'restart'
which was printed at the 15.76 ts.
The 'Fixes' tag here below points to the first pr_debug() used without
'\n' in net/mptcp. This patch could be split in many small ones, with
different Fixes tag, but it doesn't seem worth it, because it is easy to
re-generate this patch with this simple 'sed' command:
git grep -l pr_debug -- net/mptcp |
xargs sed -i "s/\(pr_debug(\".*[^n]\)\(\"[,)]\)/\1\\\n\2/g"
So in case of conflicts, simply drop the modifications, and launch this
command.
Fixes:
|
||
|
Matthieu Baerts (NGI0)
|
1511cc9611 |
mptcp: sched: check both backup in retrans
BugLink: https://bugs.launchpad.net/bugs/2084225 commit 2a1f596ebb23eadc0f9b95a8012e18ef76295fc8 upstream. The 'mptcp_subflow_context' structure has two items related to the backup flags: - 'backup': the subflow has been marked as backup by the other peer - 'request_bkup': the backup flag has been set by the host Looking only at the 'backup' flag can make sense in some cases, but it is not the behaviour of the default packet scheduler when selecting paths. As explained in the commit b6a66e521a20 ("mptcp: sched: check both directions for backup"), the packet scheduler should look at both flags, because that was the behaviour from the beginning: the 'backup' flag was set by accident instead of the 'request_bkup' one. Now that the latter has been fixed, get_retrans() needs to be adapted as well. Fixes: b6a66e521a20 ("mptcp: sched: check both directions for backup") Cc: stable@vger.kernel.org Reviewed-by: Mat Martineau <martineau@kernel.org> Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org> Link: https://patch.msgid.link/20240826-net-mptcp-close-extra-sf-fin-v1-3-905199fe1172@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Koichiro Den <koichiro.den@canonical.com> Signed-off-by: Stefan Bader <stefan.bader@canonical.com> |
||
|
Matthieu Baerts (NGI0)
|
688443c82a |
mptcp: close subflow when receiving TCP+FIN
BugLink: https://bugs.launchpad.net/bugs/2084225
commit f09b0ad55a1196f5891663f8888463c0541059cb upstream.
When a peer decides to close one subflow in the middle of a connection
having multiple subflows, the receiver of the first FIN should accept
that, and close the subflow on its side as well. If not, the subflow
will stay half closed, and would even continue to be used until the end
of the MPTCP connection or a reset from the network.
The issue has not been seen before, probably because the in-kernel
path-manager always sends a RM_ADDR before closing the subflow. Upon the
reception of this RM_ADDR, the other peer will initiate the closure on
its side as well. On the other hand, if the RM_ADDR is lost, or if the
path-manager of the other peer only closes the subflow without sending a
RM_ADDR, the subflow would switch to TCP_CLOSE_WAIT, but that's it,
leaving the subflow half-closed.
So now, when the subflow switches to the TCP_CLOSE_WAIT state, and if
the MPTCP connection has not been closed before with a DATA_FIN, the
kernel owning the subflow schedules its worker to initiate the closure
on its side as well.
This issue can be easily reproduced with packetdrill, as visible in [1],
by creating an additional subflow, injecting a FIN+ACK before sending
the DATA_FIN, and expecting a FIN+ACK in return.
Fixes:
|
||
|
Paolo Abeni
|
dd2bd28ee0 |
mptcp: fix bad RCVPRUNED mib accounting
BugLink: https://bugs.launchpad.net/bugs/2083488
commit 0a567c2a10033bf04ed618368d179bce6977984b upstream.
Since its introduction, the mentioned MIB accounted for the wrong
event: wake-up being skipped as not-needed on some edge condition
instead of incoming skb being dropped after landing in the (subflow)
receive queue.
Move the increment in the correct location.
Fixes:
|
||
|
Matthieu Baerts (NGI0)
|
7f50ecab40 |
mptcp: sched: check both directions for backup
BugLink: https://bugs.launchpad.net/bugs/2083488
commit b6a66e521a2032f7fcba2af5a9bcbaeaa19b7ca3 upstream.
The 'mptcp_subflow_context' structure has two items related to the
backup flags:
- 'backup': the subflow has been marked as backup by the other peer
- 'request_bkup': the backup flag has been set by the host
Before this patch, the scheduler was only looking at the 'backup' flag.
That can make sense in some cases, but it looks like that's not what we
wanted for the general use, because either the path-manager was setting
both of them when sending an MP_PRIO, or the receiver was duplicating
the 'backup' flag in the subflow request.
Note that the use of these two flags in the path-manager are going to be
fixed in the next commits, but this change here is needed not to modify
the behaviour.
Fixes:
|
||
 Eric Dumazet
|
6b5b7d3ec9 |
tcp: annotate data-races around tp->window_clamp
BugLink: https://bugs.launchpad.net/bugs/2083488 [ Upstream commit f410cbea9f3d2675b4c8e52af1d1985b11b387d1 ] tp->window_clamp can be read locklessly, add READ_ONCE() and WRITE_ONCE() annotations. Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Jason Xing <kerneljasonxing@gmail.com> Link: https://lore.kernel.org/r/20240404114231.2195171-1-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Stable-dep-of: 05f76b2d634e ("tcp: Adjust clamping window for applications specifying SO_RCVBUF") Signed-off-by: Sasha Levin <sashal@kernel.org> Signed-off-by: Koichiro Den <koichiro.den@canonical.com> Signed-off-by: Roxana Nicolescu <roxana.nicolescu@canonical.com> |
||
|
Paolo Abeni
|
ebea6a780f |
mptcp: ensure snd_una is properly initialized on connect
BugLink: https://bugs.launchpad.net/bugs/2074091
commit 8031b58c3a9b1db3ef68b3bd749fbee2e1e1aaa3 upstream.
This is strictly related to commit fb7a0d334894 ("mptcp: ensure snd_nxt
is properly initialized on connect"). It turns out that syzkaller can
trigger the retransmit after fallback and before processing any other
incoming packet - so that snd_una is still left uninitialized.
Address the issue explicitly initializing snd_una together with snd_nxt
and write_seq.
Suggested-by: Mat Martineau <martineau@kernel.org>
Fixes:
|
||
 Jason Xing
|
16d1b6ed9b |
mptcp: count CLOSE-WAIT sockets for MPTCP_MIB_CURRESTAB
BugLink: https://bugs.launchpad.net/bugs/2074091
[ Upstream commit 9633e9377e6af0244f7381e86b9aac5276f5be97 ]
Like previous patch does in TCP, we need to adhere to RFC 1213:
"tcpCurrEstab OBJECT-TYPE
...
The number of TCP connections for which the current state
is either ESTABLISHED or CLOSE- WAIT."
So let's consider CLOSE-WAIT sockets.
The logic of counting
When we increment the counter?
a) Only if we change the state to ESTABLISHED.
When we decrement the counter?
a) if the socket leaves ESTABLISHED and will never go into CLOSE-WAIT,
say, on the client side, changing from ESTABLISHED to FIN-WAIT-1.
b) if the socket leaves CLOSE-WAIT, say, on the server side, changing
from CLOSE-WAIT to LAST-ACK.
Fixes:
|
||
|
Paolo Abeni
|
f53275c6d9 |
mptcp: implement TCP_NOTSENT_LOWAT support
BugLink: https://bugs.launchpad.net/bugs/2071621 [ Upstream commit 29b5e5ef87397963ca38d3eec0d296ad1c979bbc ] Add support for such socket option storing the user-space provided value in a new msk field, and using such data to implement the _mptcp_stream_memory_free() helper, similar to the TCP one. To avoid adding more indirect calls in the fast path, open-code a variant of sk_stream_memory_free() in mptcp_sendmsg() and add direct calls to the mptcp stream memory free helper where possible. Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/464 Signed-off-by: Paolo Abeni <pabeni@redhat.com> Reviewed-by: Mat Martineau <martineau@kernel.org> Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net> Stable-dep-of: bd11dc4fb969 ("mptcp: fix full TCP keep-alive support") Signed-off-by: Sasha Levin <sashal@kernel.org> Signed-off-by: Manuel Diewald <manuel.diewald@canonical.com> Signed-off-by: Stefan Bader <stefan.bader@canonical.com> |
||
|
Paolo Abeni
|
ec995091f1 |
mptcp: cleanup writer wake-up
BugLink: https://bugs.launchpad.net/bugs/2071621
[ Upstream commit 037db6ea57da7a134a8183dead92d64ef92babee ]
After commit
|
||
|
Paolo Abeni
|
e5189fbb87 |
mptcp: ensure snd_nxt is properly initialized on connect
BugLink: https://bugs.launchpad.net/bugs/2070349
commit fb7a0d334894206ae35f023a82cad5a290fd7386 upstream.
Christoph reported a splat hinting at a corrupted snd_una:
WARNING: CPU: 1 PID: 38 at net/mptcp/protocol.c:1005 __mptcp_clean_una+0x4b3/0x620 net/mptcp/protocol.c:1005
Modules linked in:
CPU: 1 PID: 38 Comm: kworker/1:1 Not tainted 6.9.0-rc1-gbbeac67456c9 #59
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7 04/01/2014
Workqueue: events mptcp_worker
RIP: 0010:__mptcp_clean_una+0x4b3/0x620 net/mptcp/protocol.c:1005
Code: be 06 01 00 00 bf 06 01 00 00 e8 a8 12 e7 fe e9 00 fe ff ff e8
8e 1a e7 fe 0f b7 ab 3e 02 00 00 e9 d3 fd ff ff e8 7d 1a e7 fe
<0f> 0b 4c 8b bb e0 05 00 00 e9 74 fc ff ff e8 6a 1a e7 fe 0f 0b e9
RSP: 0018:ffffc9000013fd48 EFLAGS: 00010293
RAX: 0000000000000000 RBX: ffff8881029bd280 RCX: ffffffff82382fe4
RDX: ffff8881003cbd00 RSI: ffffffff823833c3 RDI: 0000000000000001
RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000000
R10: 0000000000000000 R11: fefefefefefefeff R12: ffff888138ba8000
R13: 0000000000000106 R14: ffff8881029bd908 R15: ffff888126560000
FS: 0000000000000000(0000) GS:ffff88813bd00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f604a5dae38 CR3: 0000000101dac002 CR4: 0000000000170ef0
Call Trace:
<TASK>
__mptcp_clean_una_wakeup net/mptcp/protocol.c:1055 [inline]
mptcp_clean_una_wakeup net/mptcp/protocol.c:1062 [inline]
__mptcp_retrans+0x7f/0x7e0 net/mptcp/protocol.c:2615
mptcp_worker+0x434/0x740 net/mptcp/protocol.c:2767
process_one_work+0x1e0/0x560 kernel/workqueue.c:3254
process_scheduled_works kernel/workqueue.c:3335 [inline]
worker_thread+0x3c7/0x640 kernel/workqueue.c:3416
kthread+0x121/0x170 kernel/kthread.c:388
ret_from_fork+0x44/0x50 arch/x86/kernel/process.c:147
ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:243
</TASK>
When fallback to TCP happens early on a client socket, snd_nxt
is not yet initialized and any incoming ack will copy such value
into snd_una. If the mptcp worker (dumbly) tries mptcp-level
re-injection after such ack, that would unconditionally trigger a send
buffer cleanup using 'bad' snd_una values.
We could easily disable re-injection for fallback sockets, but such
dumb behavior already helped catching a few subtle issues and a very
low to zero impact in practice.
Instead address the issue always initializing snd_nxt (and write_seq,
for consistency) at connect time.
Fixes:
|
||
 Davide Caratti
|
a46de6b72b |
mptcp: don't account accept() of non-MPC client as fallback to TCP
BugLink: https://bugs.launchpad.net/bugs/2065400
[ Upstream commit 7a1b3490f47e88ec4cbde65f1a77a0f4bc972282 ]
Current MPTCP servers increment MPTcpExtMPCapableFallbackACK when they
accept non-MPC connections. As reported by Christoph, this is "surprising"
because the counter might become greater than MPTcpExtMPCapableSYNRX.
MPTcpExtMPCapableFallbackACK counter's name suggests it should only be
incremented when a connection was seen using MPTCP options, then a
fallback to TCP has been done. Let's do that by incrementing it when
the subflow context of an inbound MPC connection attempt is dropped.
Also, update mptcp_connect.sh kselftest, to ensure that the
above MIB does not increment in case a pure TCP client connects to a
MPTCP server.
Fixes:
|
||
|
Davide Caratti
|
10048689de |
mptcp: fix double-free on socket dismantle
when MPTCP server accepts an incoming connection, it clones its listener
socket. However, the pointer to 'inet_opt' for the new socket has the same
value as the original one: as a consequence, on program exit it's possible
to observe the following splat:
BUG: KASAN: double-free in inet_sock_destruct+0x54f/0x8b0
Free of addr ffff888485950880 by task swapper/25/0
CPU: 25 PID: 0 Comm: swapper/25 Kdump: loaded Not tainted 6.8.0-rc1+ #609
Hardware name: Supermicro SYS-6027R-72RF/X9DRH-7TF/7F/iTF/iF, BIOS 3.0 07/26/2013
Call Trace:
<IRQ>
dump_stack_lvl+0x32/0x50
print_report+0xca/0x620
kasan_report_invalid_free+0x64/0x90
__kasan_slab_free+0x1aa/0x1f0
kfree+0xed/0x2e0
inet_sock_destruct+0x54f/0x8b0
__sk_destruct+0x48/0x5b0
rcu_do_batch+0x34e/0xd90
rcu_core+0x559/0xac0
__do_softirq+0x183/0x5a4
irq_exit_rcu+0x12d/0x170
sysvec_apic_timer_interrupt+0x6b/0x80
</IRQ>
<TASK>
asm_sysvec_apic_timer_interrupt+0x16/0x20
RIP: 0010:cpuidle_enter_state+0x175/0x300
Code: 30 00 0f 84 1f 01 00 00 83 e8 01 83 f8 ff 75 e5 48 83 c4 18 44 89 e8 5b 5d 41 5c 41 5d 41 5e 41 5f c3 cc cc cc cc fb 45 85 ed <0f> 89 60 ff ff ff 48 c1 e5 06 48 c7 43 18 00 00 00 00 48 83 44 2b
RSP: 0018:ffff888481cf7d90 EFLAGS: 00000202
RAX: 0000000000000000 RBX: ffff88887facddc8 RCX: 0000000000000000
RDX: 1ffff1110ff588b1 RSI: 0000000000000019 RDI: ffff88887fac4588
RBP: 0000000000000004 R08: 0000000000000002 R09: 0000000000043080
R10: 0009b02ea273363f R11: ffff88887fabf42b R12: ffffffff932592e0
R13: 0000000000000004 R14: 0000000000000000 R15: 00000022c880ec80
cpuidle_enter+0x4a/0xa0
do_idle+0x310/0x410
cpu_startup_entry+0x51/0x60
start_secondary+0x211/0x270
secondary_startup_64_no_verify+0x184/0x18b
</TASK>
Allocated by task 6853:
kasan_save_stack+0x1c/0x40
kasan_save_track+0x10/0x30
__kasan_kmalloc+0xa6/0xb0
__kmalloc+0x1eb/0x450
cipso_v4_sock_setattr+0x96/0x360
netlbl_sock_setattr+0x132/0x1f0
selinux_netlbl_socket_post_create+0x6c/0x110
selinux_socket_post_create+0x37b/0x7f0
security_socket_post_create+0x63/0xb0
__sock_create+0x305/0x450
__sys_socket_create.part.23+0xbd/0x130
__sys_socket+0x37/0xb0
__x64_sys_socket+0x6f/0xb0
do_syscall_64+0x83/0x160
entry_SYSCALL_64_after_hwframe+0x6e/0x76
Freed by task 6858:
kasan_save_stack+0x1c/0x40
kasan_save_track+0x10/0x30
kasan_save_free_info+0x3b/0x60
__kasan_slab_free+0x12c/0x1f0
kfree+0xed/0x2e0
inet_sock_destruct+0x54f/0x8b0
__sk_destruct+0x48/0x5b0
subflow_ulp_release+0x1f0/0x250
tcp_cleanup_ulp+0x6e/0x110
tcp_v4_destroy_sock+0x5a/0x3a0
inet_csk_destroy_sock+0x135/0x390
tcp_fin+0x416/0x5c0
tcp_data_queue+0x1bc8/0x4310
tcp_rcv_state_process+0x15a3/0x47b0
tcp_v4_do_rcv+0x2c1/0x990
tcp_v4_rcv+0x41fb/0x5ed0
ip_protocol_deliver_rcu+0x6d/0x9f0
ip_local_deliver_finish+0x278/0x360
ip_local_deliver+0x182/0x2c0
ip_rcv+0xb5/0x1c0
__netif_receive_skb_one_core+0x16e/0x1b0
process_backlog+0x1e3/0x650
__napi_poll+0xa6/0x500
net_rx_action+0x740/0xbb0
__do_softirq+0x183/0x5a4
The buggy address belongs to the object at ffff888485950880
which belongs to the cache kmalloc-64 of size 64
The buggy address is located 0 bytes inside of
64-byte region [ffff888485950880, ffff8884859508c0)
The buggy address belongs to the physical page:
page:0000000056d1e95e refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff888485950700 pfn:0x485950
flags: 0x57ffffc0000800(slab|node=1|zone=2|lastcpupid=0x1fffff)
page_type: 0xffffffff()
raw: 0057ffffc0000800 ffff88810004c640 ffffea00121b8ac0 dead000000000006
raw: ffff888485950700 0000000000200019 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected
Memory state around the buggy address:
ffff888485950780: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc
ffff888485950800: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc
>ffff888485950880: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc
^
ffff888485950900: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc
ffff888485950980: 00 00 00 00 00 01 fc fc fc fc fc fc fc fc fc fc
Something similar (a refcount underflow) happens with CALIPSO/IPv6. Fix
this by duplicating IP / IPv6 options after clone, so that
ip{,6}_sock_destruct() doesn't end up freeing the same memory area twice.
Fixes:
|
||
|
Paolo Abeni
|
adf1bb78da |
mptcp: fix snd_wnd initialization for passive socket
Such value should be inherited from the first subflow, but
passive sockets always used 'rsk_rcv_wnd'.
Fixes:
|
||
|
Paolo Abeni
|
b9cd26f640 |
mptcp: push at DSS boundaries
when inserting not contiguous data in the subflow write queue,
the protocol creates a new skb and prevent the TCP stack from
merging it later with already queued skbs by setting the EOR marker.
Still no push flag is explicitly set at the end of previous GSO
packet, making the aggregation on the receiver side sub-optimal -
and packetdrill self-tests less predictable.
Explicitly mark the end of not contiguous DSS with the push flag.
Fixes:
|
||
|
Paolo Abeni
|
a7cfe77663 |
mptcp: fix data races on local_id
The local address id is accessed lockless by the NL PM, add
all the required ONCE annotation. There is a caveat: the local
id can be initialized late in the subflow life-cycle, and its
validity is controlled by the local_id_valid flag.
Remove such flag and encode the validity in the local_id field
itself with negative value before initialization. That allows
accessing the field consistently with a single read operation.
Fixes:
|
||
|
Paolo Abeni
|
e4a0fa47e8 |
mptcp: corner case locking for rx path fields initialization
Most MPTCP-level related fields are under the mptcp data lock
protection, but are written one-off without such lock at MPC
complete time, both for the client and the server
Leverage the mptcp_propagate_state() infrastructure to move such
initialization under the proper lock client-wise.
The server side critical init steps are done by
mptcp_subflow_fully_established(): ensure the caller properly held the
relevant lock, and avoid acquiring the same lock in the nested scopes.
There are no real potential races, as write access to such fields
is implicitly serialized by the MPTCP state machine; the primary
goal is consistency.
Fixes:
|
||
|
Paolo Abeni
|
3f83d8a77e |
mptcp: fix more tx path fields initialization
The 'msk->write_seq' and 'msk->snd_nxt' are always updated under
the msk socket lock, except at MPC handshake completiont time.
Builds-up on the previous commit to move such init under the relevant
lock.
There are no known problems caused by the potential race, the
primary goal is consistency.
Fixes:
|
||
|
Paolo Abeni
|
013e3179db |
mptcp: fix rcv space initialization
mptcp_rcv_space_init() is supposed to happen under the msk socket
lock, but active msk socket does that without such protection.
Leverage the existing mptcp_propagate_state() helper to that extent.
We need to ensure mptcp_rcv_space_init will happen before
mptcp_rcv_space_adjust(), and the release_cb does not assure that:
explicitly check for such condition.
While at it, move the wnd_end initialization out of mptcp_rcv_space_init(),
it never belonged there.
Note that the race does not produce ill effect in practice, but
change allows cleaning-up and defying better the locking model.
Fixes:
|
||
|
Paolo Abeni
|
bdd70eb689 |
mptcp: drop the push_pending field
Such field is there to avoid acquiring the data lock in a few spots,
but it adds complexity to the already non trivial locking schema.
All the relevant call sites (mptcp-level re-injection, set socket
options), are slow-path, drop such field in favor of 'cb_flags', adding
the relevant locking.
This patch could be seen as an improvement, instead of a fix. But it
simplifies the next patch. The 'Fixes' tag has been added to help having
this series backported to stable.
Fixes:
|
||
|
Paolo Abeni
|
b6c620dc43 |
mptcp: fix data re-injection from stale subflow
When the MPTCP PM detects that a subflow is stale, all the packet scheduler must re-inject all the mptcp-level unacked data. To avoid acquiring unneeded locks, it first try to check if any unacked data is present at all in the RTX queue, but such check is currently broken, as it uses TCP-specific helper on an MPTCP socket. Funnily enough fuzzers and static checkers are happy, as the accessed memory still belongs to the mptcp_sock struct, and even from a functional perspective the recovery completed successfully, as the short-cut test always failed. A recent unrelated TCP change - commit |
||
 Geliang Tang
|
c693a85164 |
mptcp: use mptcp_set_state
This patch replaces all the 'inet_sk_state_store()' calls under net/mptcp with the new helper mptcp_set_state(). Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/460 Signed-off-by: Geliang Tang <geliang.tang@linux.dev> Acked-by: Paolo Abeni <pabeni@redhat.com> Reviewed-by: Matthieu Baerts <matttbe@kernel.org> Signed-off-by: Matthieu Baerts <matttbe@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net> |
||
|
Geliang Tang
|
d9cd27b8cd |
mptcp: add CurrEstab MIB counter support
Add a new MIB counter named MPTCP_MIB_CURRESTAB to count current established MPTCP connections, similar to TCP_MIB_CURRESTAB. This is useful to quickly list the number of MPTCP connections without having to iterate over all of them. This patch adds a new helper function mptcp_set_state(): if the state switches from or to ESTABLISHED state, this newly added counter is incremented. This helper is going to be used in the following patch. Similar to MPTCP_INC_STATS(), a new helper called MPTCP_DEC_STATS() is also needed to decrement a MIB counter. Signed-off-by: Geliang Tang <geliang.tang@linux.dev> Acked-by: Paolo Abeni <pabeni@redhat.com> Reviewed-by: Matthieu Baerts <matttbe@kernel.org> Signed-off-by: Matthieu Baerts <matttbe@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net> |
||
|
Davide Caratti
|
8e2b8a9fa5 |
mptcp: don't overwrite sock_ops in mptcp_is_tcpsk()
Eric Dumazet suggests: > The fact that mptcp_is_tcpsk() was able to write over sock->ops was a > bit strange to me. > mptcp_is_tcpsk() should answer a question, with a read-only argument. re-factor code to avoid overwriting sock_ops inside that function. Also, change the helper name to reflect the semantics and to disambiguate from its dual, sk_is_mptcp(). While at it, collapse mptcp_stream_accept() and mptcp_accept() into a single function, where fallback / non-fallback are separated into a single sk_is_mptcp() conditional. Link: https://github.com/multipath-tcp/mptcp_net-next/issues/432 Suggested-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Davide Caratti <dcaratti@redhat.com> Acked-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Matthieu Baerts <matttbe@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net> |
||
|
Paolo Abeni
|
4fd19a3070 |
mptcp: fix inconsistent state on fastopen race
The netlink PM can race with fastopen self-connect attempts, shutting
down the first subflow via:
MPTCP_PM_CMD_DEL_ADDR -> mptcp_nl_remove_id_zero_address ->
mptcp_pm_nl_rm_subflow_received -> mptcp_close_ssk
and transitioning such subflow to FIN_WAIT1 status before the syn-ack
packet is processed. The MPTCP code does not react to such state change,
leaving the connection in not-fallback status and the subflow handshake
uncompleted, triggering the following splat:
WARNING: CPU: 0 PID: 10630 at net/mptcp/subflow.c:1405 subflow_data_ready+0x39f/0x690 net/mptcp/subflow.c:1405
Modules linked in:
CPU: 0 PID: 10630 Comm: kworker/u4:11 Not tainted 6.6.0-syzkaller-14500-g1c41041124bd #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/09/2023
Workqueue: bat_events batadv_nc_worker
RIP: 0010:subflow_data_ready+0x39f/0x690 net/mptcp/subflow.c:1405
Code: 18 89 ee e8 e3 d2 21 f7 40 84 ed 75 1f e8 a9 d7 21 f7 44 89 fe bf 07 00 00 00 e8 0c d3 21 f7 41 83 ff 07 74 07 e8 91 d7 21 f7 <0f> 0b e8 8a d7 21 f7 48 89 df e8 d2 b2 ff ff 31 ff 89 c5 89 c6 e8
RSP: 0018:ffffc90000007448 EFLAGS: 00010246
RAX: 0000000000000000 RBX: ffff888031efc700 RCX: ffffffff8a65baf4
RDX: ffff888043222140 RSI: ffffffff8a65baff RDI: 0000000000000005
RBP: 0000000000000000 R08: 0000000000000005 R09: 0000000000000007
R10: 000000000000000b R11: 0000000000000000 R12: 1ffff92000000e89
R13: ffff88807a534d80 R14: ffff888021c11a00 R15: 000000000000000b
FS: 0000000000000000(0000) GS:ffff8880b9800000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007fa19a0ffc81 CR3: 000000007a2db000 CR4: 00000000003506f0
DR0: 000000000000d8dd DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Call Trace:
<IRQ>
tcp_data_ready+0x14c/0x5b0 net/ipv4/tcp_input.c:5128
tcp_data_queue+0x19c3/0x5190 net/ipv4/tcp_input.c:5208
tcp_rcv_state_process+0x11ef/0x4e10 net/ipv4/tcp_input.c:6844
tcp_v4_do_rcv+0x369/0xa10 net/ipv4/tcp_ipv4.c:1929
tcp_v4_rcv+0x3888/0x3b30 net/ipv4/tcp_ipv4.c:2329
ip_protocol_deliver_rcu+0x9f/0x480 net/ipv4/ip_input.c:205
ip_local_deliver_finish+0x2e4/0x510 net/ipv4/ip_input.c:233
NF_HOOK include/linux/netfilter.h:314 [inline]
NF_HOOK include/linux/netfilter.h:308 [inline]
ip_local_deliver+0x1b6/0x550 net/ipv4/ip_input.c:254
dst_input include/net/dst.h:461 [inline]
ip_rcv_finish+0x1c4/0x2e0 net/ipv4/ip_input.c:449
NF_HOOK include/linux/netfilter.h:314 [inline]
NF_HOOK include/linux/netfilter.h:308 [inline]
ip_rcv+0xce/0x440 net/ipv4/ip_input.c:569
__netif_receive_skb_one_core+0x115/0x180 net/core/dev.c:5527
__netif_receive_skb+0x1f/0x1b0 net/core/dev.c:5641
process_backlog+0x101/0x6b0 net/core/dev.c:5969
__napi_poll.constprop.0+0xb4/0x540 net/core/dev.c:6531
napi_poll net/core/dev.c:6600 [inline]
net_rx_action+0x956/0xe90 net/core/dev.c:6733
__do_softirq+0x21a/0x968 kernel/softirq.c:553
do_softirq kernel/softirq.c:454 [inline]
do_softirq+0xaa/0xe0 kernel/softirq.c:441
</IRQ>
<TASK>
__local_bh_enable_ip+0xf8/0x120 kernel/softirq.c:381
spin_unlock_bh include/linux/spinlock.h:396 [inline]
batadv_nc_purge_paths+0x1ce/0x3c0 net/batman-adv/network-coding.c:471
batadv_nc_worker+0x9b1/0x10e0 net/batman-adv/network-coding.c:722
process_one_work+0x884/0x15c0 kernel/workqueue.c:2630
process_scheduled_works kernel/workqueue.c:2703 [inline]
worker_thread+0x8b9/0x1290 kernel/workqueue.c:2784
kthread+0x33c/0x440 kernel/kthread.c:388
ret_from_fork+0x45/0x80 arch/x86/kernel/process.c:147
ret_from_fork_asm+0x11/0x20 arch/x86/entry/entry_64.S:242
</TASK>
To address the issue, catch the racing subflow state change and
use it to cause the MPTCP fallback. Such fallback is also used to
cause the first subflow state propagation to the msk socket via
mptcp_set_connected(). After this change, the first subflow can
additionally propagate the TCP_FIN_WAIT1 state, so rename the
helper accordingly.
Finally, if the state propagation is delayed to the msk release
callback, the first subflow can change to a different state in between.
Cache the relevant target state in a new msk-level field and use
such value to update the msk state at release time.
Fixes:
|
||
|
Paolo Abeni
|
d109a77672 |
mptcp: fix possible NULL pointer dereference on close
After the blamed commit below, the MPTCP release callback can
dereference the first subflow pointer via __mptcp_set_connected()
and send buffer auto-tuning. Such pointer is always expected to be
valid, except at socket destruction time, when the first subflow is
deleted and the pointer zeroed.
If the connect event is handled by the release callback while the
msk socket is finally released, MPTCP hits the following splat:
general protection fault, probably for non-canonical address 0xdffffc00000000f2: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000790-0x0000000000000797]
CPU: 1 PID: 26719 Comm: syz-executor.2 Not tainted 6.6.0-syzkaller-10102-gff269e2cd5ad #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/09/2023
RIP: 0010:mptcp_subflow_ctx net/mptcp/protocol.h:542 [inline]
RIP: 0010:__mptcp_propagate_sndbuf net/mptcp/protocol.h:813 [inline]
RIP: 0010:__mptcp_set_connected+0x57/0x3e0 net/mptcp/subflow.c:424
RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffffffff8a62323c
RDX: 00000000000000f2 RSI: ffffffff8a630116 RDI: 0000000000000790
RBP: ffff88803334b100 R08: 0000000000000001 R09: 0000000000000000
R10: 0000000000000001 R11: 0000000000000034 R12: ffff88803334b198
R13: ffff888054f0b018 R14: 0000000000000000 R15: ffff88803334b100
FS: 0000000000000000(0000) GS:ffff8880b9900000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007fbcb4f75198 CR3: 000000006afb5000 CR4: 00000000003506f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
mptcp_release_cb+0xa2c/0xc40 net/mptcp/protocol.c:3405
release_sock+0xba/0x1f0 net/core/sock.c:3537
mptcp_close+0x32/0xf0 net/mptcp/protocol.c:3084
inet_release+0x132/0x270 net/ipv4/af_inet.c:433
inet6_release+0x4f/0x70 net/ipv6/af_inet6.c:485
__sock_release+0xae/0x260 net/socket.c:659
sock_close+0x1c/0x20 net/socket.c:1419
__fput+0x270/0xbb0 fs/file_table.c:394
task_work_run+0x14d/0x240 kernel/task_work.c:180
exit_task_work include/linux/task_work.h:38 [inline]
do_exit+0xa92/0x2a20 kernel/exit.c:876
do_group_exit+0xd4/0x2a0 kernel/exit.c:1026
get_signal+0x23ba/0x2790 kernel/signal.c:2900
arch_do_signal_or_restart+0x90/0x7f0 arch/x86/kernel/signal.c:309
exit_to_user_mode_loop kernel/entry/common.c:168 [inline]
exit_to_user_mode_prepare+0x11f/0x240 kernel/entry/common.c:204
__syscall_exit_to_user_mode_work kernel/entry/common.c:285 [inline]
syscall_exit_to_user_mode+0x1d/0x60 kernel/entry/common.c:296
do_syscall_64+0x4b/0x110 arch/x86/entry/common.c:88
entry_SYSCALL_64_after_hwframe+0x63/0x6b
RIP: 0033:0x7fb515e7cae9
Code: Unable to access opcode bytes at 0x7fb515e7cabf.
RSP: 002b:00007fb516c560c8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
RAX: 000000000000003c RBX: 00007fb515f9c120 RCX: 00007fb515e7cae9
RDX: 0000000000000000 RSI: 0000000020000140 RDI: 0000000000000006
RBP: 00007fb515ec847a R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 000000000000006e R14: 00007fb515f9c120 R15: 00007ffc631eb968
</TASK>
To avoid sparkling unneeded conditionals, address the issue explicitly
checking msk->first only in the critical place.
Fixes:
|
||
|
Paolo Abeni
|
9fce92f050 |
mptcp: deal with large GSO size
After the blamed commit below, the TCP sockets (and the MPTCP subflows)
can build egress packets larger than 64K. That exceeds the maximum DSS
data size, the length being misrepresent on the wire and the stream being
corrupted, as later observed on the receiver:
WARNING: CPU: 0 PID: 9696 at net/mptcp/protocol.c:705 __mptcp_move_skbs_from_subflow+0x2604/0x26e0
CPU: 0 PID: 9696 Comm: syz-executor.7 Not tainted 6.6.0-rc5-gcd8bdf563d46 #45
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7 04/01/2014
netlink: 8 bytes leftover after parsing attributes in process `syz-executor.4'.
RIP: 0010:__mptcp_move_skbs_from_subflow+0x2604/0x26e0 net/mptcp/protocol.c:705
RSP: 0018:ffffc90000006e80 EFLAGS: 00010246
RAX: ffffffff83e9f674 RBX: ffff88802f45d870 RCX: ffff888102ad0000
netlink: 8 bytes leftover after parsing attributes in process `syz-executor.4'.
RDX: 0000000080000303 RSI: 0000000000013908 RDI: 0000000000003908
RBP: ffffc90000007110 R08: ffffffff83e9e078 R09: 1ffff1100e548c8a
R10: dffffc0000000000 R11: ffffed100e548c8b R12: 0000000000013908
R13: dffffc0000000000 R14: 0000000000003908 R15: 000000000031cf29
FS: 00007f239c47e700(0000) GS:ffff88811b200000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f239c45cd78 CR3: 000000006a66c006 CR4: 0000000000770ef0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600
PKRU: 55555554
Call Trace:
<IRQ>
mptcp_data_ready+0x263/0xac0 net/mptcp/protocol.c:819
subflow_data_ready+0x268/0x6d0 net/mptcp/subflow.c:1409
tcp_data_queue+0x21a1/0x7a60 net/ipv4/tcp_input.c:5151
tcp_rcv_established+0x950/0x1d90 net/ipv4/tcp_input.c:6098
tcp_v6_do_rcv+0x554/0x12f0 net/ipv6/tcp_ipv6.c:1483
tcp_v6_rcv+0x2e26/0x3810 net/ipv6/tcp_ipv6.c:1749
ip6_protocol_deliver_rcu+0xd6b/0x1ae0 net/ipv6/ip6_input.c:438
ip6_input+0x1c5/0x470 net/ipv6/ip6_input.c:483
ipv6_rcv+0xef/0x2c0 include/linux/netfilter.h:304
__netif_receive_skb+0x1ea/0x6a0 net/core/dev.c:5532
process_backlog+0x353/0x660 net/core/dev.c:5974
__napi_poll+0xc6/0x5a0 net/core/dev.c:6536
net_rx_action+0x6a0/0xfd0 net/core/dev.c:6603
__do_softirq+0x184/0x524 kernel/softirq.c:553
do_softirq+0xdd/0x130 kernel/softirq.c:454
Address the issue explicitly bounding the maximum GSO size to what MPTCP
actually allows.
Reported-by: Christoph Paasch <cpaasch@apple.com>
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/450
Fixes:
|
||
 Geliang Tang
|
a16c054b52 |
mptcp: use mptcp_get_ext helper
Use mptcp_get_ext() helper defined in protocol.h instead of open-coding it in mptcp_sendmsg_frag(). Reviewed-by: Matthieu Baerts <matttbe@kernel.org> Signed-off-by: Geliang Tang <geliang.tang@suse.com> Signed-off-by: Mat Martineau <martineau@kernel.org> Link: https://lore.kernel.org/r/20231025-send-net-next-20231025-v1-6-db8f25f798eb@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> |
||
|
Geliang Tang
|
74cbb0c65b |
mptcp: drop useless ssk in pm_subflow_check_next
The code using 'ssk' parameter of mptcp_pm_subflow_check_next() has been dropped in commit "95d686517884 (mptcp: fix subflow accounting on close)". So drop this useless parameter ssk. Reviewed-by: Matthieu Baerts <matttbe@kernel.org> Signed-off-by: Geliang Tang <geliang.tang@suse.com> Signed-off-by: Mat Martineau <martineau@kernel.org> Link: https://lore.kernel.org/r/20231025-send-net-next-20231025-v1-4-db8f25f798eb@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> |
||
|
Paolo Abeni
|
8005184fd1 |
mptcp: refactor sndbuf auto-tuning
The MPTCP protocol account for the data enqueued on all the subflows to the main socket send buffer, while the send buffer auto-tuning algorithm set the main socket send buffer size as the max size among the subflows. That causes bad performances when at least one subflow is sndbuf limited, e.g. due to very high latency, as the MPTCP scheduler can't even fill such buffer. Change the send-buffer auto-tuning algorithm to compute the main socket send buffer size as the sum of all the subflows buffer size. 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/20231023-send-net-next-20231023-2-v1-9-9dc60939d371@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> |
||
|
Paolo Abeni
|
a1ab24e5fc |
mptcp: consolidate sockopt synchronization
Move the socket option synchronization for active subflows at subflow creation time. This allows removing the now unused unlocked variant of such helper. While at that, clean-up a bit the mptcp_subflow_create_socket() errors path. 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/20231023-send-net-next-20231023-2-v1-7-9dc60939d371@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> |
||
|
Paolo Abeni
|
0ffe8e7490 |
mptcp: use copy_from_iter helpers on transmit
The perf traces show an high cost for the MPTCP transmit path memcpy. It turn out that the helper currently in use carries quite a bit of unneeded overhead, e.g. to map/unmap the memory pages. Moving to the 'copy_from_iter' variant removes such overhead and additionally gains the no-cache support. 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/20231023-send-net-next-20231023-2-v1-6-9dc60939d371@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> |
||
|
Paolo Abeni
|
5684ab1a0e |
mptcp: give rcvlowat some love
The MPTCP protocol allow setting sk_rcvlowat, but the value there is currently ignored. Additionally, the default subflows sk_rcvlowat basically disables per subflow delayed ack: the MPTCP protocol move the incoming data from the subflows into the msk socket as soon as the TCP stacks invokes the subflow data_ready callback. Later, when __tcp_ack_snd_check() takes action, the subflow-level copied_seq matches rcv_nxt, and that mandate for an immediate ack. Let the mptcp receive path be aware of such threshold, explicitly tracking the amount of data available to be ready and checking vs sk_rcvlowat in mptcp_poll() and before waking-up readers. Additionally implement the set_rcvlowat() callback, to properly handle the rcvbuf auto-tuning on sk_rcvlowat changes. Finally to properly handle delayed ack, force the subflow level threshold to 0 and instead explicitly ask for an immediate ack when the msk level th is not reached. 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/20231023-send-net-next-20231023-2-v1-5-9dc60939d371@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> |
||
|
Paolo Abeni
|
d866ae9aaa |
mptcp: add a new sysctl for make after break timeout
The MPTCP protocol allows sockets with no alive subflows to stay in ESTABLISHED status for and user-defined timeout, to allow for later subflows creation. Currently such timeout is constant - TCP_TIMEWAIT_LEN. Let the user-space configure them via a newly added sysctl, to better cope with busy servers and simplify (make them faster) the relevant pktdrill tests. Note that the new know does not apply to orphaned MPTCP socket waiting for the data_fin handshake completion: they always wait TCP_TIMEWAIT_LEN. 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/20231023-send-net-next-20231023-2-v1-1-9dc60939d371@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> |
||
|
Geliang Tang
|
14c56686a6 |
mptcp: avoid sending RST when closing the initial subflow
When closing the first subflow, the MPTCP protocol unconditionally
calls tcp_disconnect(), which in turn generates a reset if the subflow
is established.
That is unexpected and different from what MPTCP does with MPJ
subflows, where resets are generated only on FASTCLOSE and other edge
scenarios.
We can't reuse for the first subflow the same code in place for MPJ
subflows, as MPTCP clean them up completely via a tcp_close() call,
while must keep the first subflow socket alive for later re-usage, due
to implementation constraints.
This patch adds a new helper __mptcp_subflow_disconnect() that
encapsulates, a logic similar to tcp_close, issuing a reset only when
the MPTCP_CF_FASTCLOSE flag is set, and performing a clean shutdown
otherwise.
Fixes:
|
||
|
Paolo Abeni
|
72377ab2d6 |
mptcp: more conservative check for zero probes
Christoph reported that the MPTCP protocol can find the subflow-level
write queue unexpectedly not empty while crafting a zero-window probe,
hitting a warning:
------------[ cut here ]------------
WARNING: CPU: 0 PID: 188 at net/mptcp/protocol.c:1312 mptcp_sendmsg_frag+0xc06/0xe70
Modules linked in:
CPU: 0 PID: 188 Comm: kworker/0:2 Not tainted 6.6.0-rc2-g1176aa719d7a #47
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7 04/01/2014
Workqueue: events mptcp_worker
RIP: 0010:mptcp_sendmsg_frag+0xc06/0xe70 net/mptcp/protocol.c:1312
RAX: 47d0530de347ff6a RBX: 47d0530de347ff6b RCX: ffff8881015d3c00
RDX: ffff8881015d3c00 RSI: 47d0530de347ff6b RDI: 47d0530de347ff6b
RBP: 47d0530de347ff6b R08: ffffffff8243c6a8 R09: ffffffff82042d9c
R10: 0000000000000002 R11: ffffffff82056850 R12: ffff88812a13d580
R13: 0000000000000001 R14: ffff88812b375e50 R15: ffff88812bbf3200
FS: 0000000000000000(0000) GS:ffff88813bc00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000695118 CR3: 0000000115dfc001 CR4: 0000000000170ef0
Call Trace:
<TASK>
__subflow_push_pending+0xa4/0x420 net/mptcp/protocol.c:1545
__mptcp_push_pending+0x128/0x3b0 net/mptcp/protocol.c:1614
mptcp_release_cb+0x218/0x5b0 net/mptcp/protocol.c:3391
release_sock+0xf6/0x100 net/core/sock.c:3521
mptcp_worker+0x6e8/0x8f0 net/mptcp/protocol.c:2746
process_scheduled_works+0x341/0x690 kernel/workqueue.c:2630
worker_thread+0x3a7/0x610 kernel/workqueue.c:2784
kthread+0x143/0x180 kernel/kthread.c:388
ret_from_fork+0x4d/0x60 arch/x86/kernel/process.c:147
ret_from_fork_asm+0x1b/0x30 arch/x86/entry/entry_64.S:304
</TASK>
The root cause of the issue is that expectations are wrong: e.g. due
to MPTCP-level re-injection we can hit the critical condition.
Explicitly avoid the zero-window probe when the subflow write queue
is not empty and drop the related warnings.
Reported-by: Christoph Paasch <cpaasch@apple.com>
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/444
Fixes:
|
||
|
Paolo Abeni
|
419ce133ab |
tcp: allow again tcp_disconnect() when threads are waiting
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:
|
||
|
Paolo Abeni
|
a5efdbcece |
mptcp: fix delegated action races
The delegated action infrastructure is prone to the following
race: different CPUs can try to schedule different delegated
actions on the same subflow at the same time.
Each of them will check different bits via mptcp_subflow_delegate(),
and will try to schedule the action on the related per-cpu napi
instance.
Depending on the timing, both can observe an empty delegated list
node, causing the same entry to be added simultaneously on two different
lists.
The root cause is that the delegated actions infra does not provide
a single synchronization point. Address the issue reserving an additional
bit to mark the subflow as scheduled for delegation. Acquiring such bit
guarantee the caller to own the delegated list node, and being able to
safely schedule the subflow.
Clear such bit only when the subflow scheduling is completed, ensuring
proper barrier in place.
Additionally swap the meaning of the delegated_action bitmask, to allow
the usage of the existing helper to set multiple bit at once.
Fixes:
|
||
|
Paolo Abeni
|
27e5ccc2d5 |
mptcp: fix dangling connection hang-up
According to RFC 8684 section 3.3: A connection is not closed unless [...] or an implementation-specific connection-level send timeout. Currently the MPTCP protocol does not implement such timeout, and connection timing-out at the TCP-level never move to close state. Introduces a catch-up condition at subflow close time to move the MPTCP socket to close, too. That additionally allows removing similar existing inside the worker. Finally, allow some additional timeout for plain ESTABLISHED mptcp sockets, as the protocol allows creating new subflows even at that point and making the connection functional again. This issue is actually present since the beginning, but it is basically impossible to solve without a long chain of functional pre-requisites topped by commit |