Age | Commit message (Collapse) | Author |
|
Syzkaller with UBSAN uncovered a scenario where a large number of
DATA_FIN retransmits caused a shift-out-of-bounds in the DATA_FIN
timeout calculation:
================================================================================
UBSAN: shift-out-of-bounds in net/mptcp/protocol.c:470:29
shift exponent 32 is too large for 32-bit type 'unsigned int'
CPU: 1 PID: 13059 Comm: kworker/1:0 Not tainted 5.17.0-rc2-00630-g5fbf21c90c60 #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014
Workqueue: events mptcp_worker
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106
ubsan_epilogue+0xb/0x5a lib/ubsan.c:151
__ubsan_handle_shift_out_of_bounds.cold+0xb2/0x20e lib/ubsan.c:330
mptcp_set_datafin_timeout net/mptcp/protocol.c:470 [inline]
__mptcp_retrans.cold+0x72/0x77 net/mptcp/protocol.c:2445
mptcp_worker+0x58a/0xa70 net/mptcp/protocol.c:2528
process_one_work+0x9df/0x16d0 kernel/workqueue.c:2307
worker_thread+0x95/0xe10 kernel/workqueue.c:2454
kthread+0x2f4/0x3b0 kernel/kthread.c:377
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
</TASK>
================================================================================
This change limits the maximum timeout by limiting the size of the
shift, which keeps all intermediate values in-bounds.
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/259
Fixes: 6477dd39e62c ("mptcp: Retransmit DATA_FIN")
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The MPTCP SIOCOUTQ implementation is not very accurate in
case of fallback: it only measures the data in the MPTCP-level
write queue, but it does not take in account the subflow
write queue utilization. In case of fallback the first can be
empty, while the latter is not.
The above produces sporadic self-tests issues and can foul
legit user-space application.
Fix the issue additionally querying the subflow in case of fallback.
Fixes: 644807e3e462 ("mptcp: add SIOCINQ, OUTQ and OUTQNSD ioctls")
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/260
Reported-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The MPTCP in kernel path manager has some constraints on incoming
addresses announce processing, so that in edge scenarios it can
end-up dropping (ignoring) some of such announces.
The above is not very limiting in practice since such scenarios are
very uncommon and MPTCP will recover due to ADD_ADDR retransmissions.
This patch adds a few MIB counters to account for such drop events
to allow easier introspection of the critical scenarios.
Fixes: f7efc7771eac ("mptcp: drop argument port from mptcp_pm_announce_addr")
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
If an MPTCP endpoint received multiple consecutive incoming
ADD_ADDR options, mptcp_pm_add_addr_received() can overwrite
the current remote address value after the PM lock is released
in mptcp_pm_nl_add_addr_received() and before such address
is echoed.
Fix the issue caching the remote address value a little earlier
and always using the cached value after releasing the PM lock.
Fixes: f7efc7771eac ("mptcp: drop argument port from mptcp_pm_announce_addr")
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
After commit a88c9e496937 ("mptcp: do not block subflows
creation on errors"), if a signal address races with a failing
subflow creation, the subflow creation failure control path
can trigger the selection of the next address to be announced
while the current announced is still pending.
The above will cause the unintended suppression of the ADD_ADDR
announce.
Fix the issue skipping the to-be-suppressed announce before it
will mark an endpoint as already used. The relevant announce
will be triggered again when the current one will complete.
Fixes: a88c9e496937 ("mptcp: do not block subflows creation on errors")
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This change updates mptcp_pm_nl_create_listen_socket() to create
listening sockets bound to IPv6 addresses (where IPv6 is supported).
Fixes: 1729cf186d8a ("mptcp: create the listening socket for new port")
Acked-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Kishen Maloor <kishen.maloor@intel.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
In preparation for FORTIFY_SOURCE performing compile-time and run-time
field bounds checking for memcpy(), memmove(), and memset(), avoid
intentionally writing across neighboring fields.
Use struct_group() to capture the fields to be reset, so that memset()
can be appropriately bounds-checked by the compiler.
Cc: Matthieu Baerts <matthieu.baerts@tessares.net>
Cc: mptcp@lists.linux.dev
Signed-off-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Link: https://lore.kernel.org/r/20220121073935.1154263-1-keescook@chromium.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
In mptcp_pm_nl_rm_addr_or_subflow(), the bit of rm_list->ids[i] in the
id_avail_bitmap should be set, not rm_list->ids[1]. This patch fixed it.
Fixes: 86e39e04482b ("mptcp: keep track of local endpoint still available for each msk")
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The MPTCP endpoint list is under RCU protection, guarded by the
pernet spinlock. mptcp_nl_cmd_set_flags() traverses the list
without acquiring the spin-lock nor under the RCU critical section.
This change addresses the issue performing the lookup and the endpoint
update under the pernet spinlock.
Fixes: 0f9f696a502e ("mptcp: add set_flags command in PM netlink")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Merge in fixes directly in prep for the 5.17 merge window.
No conflicts.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
This patch reused __mptcp_make_csum() in validate_data_csum() instead of
open-coding.
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
This patch changed the type of the last parameter of __mptcp_make_csum()
from __sum16 to __wsum. And export this function in protocol.h.
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
syzbot found a page counter underflow that was triggered by MPTCP's
reclaim code:
page_counter underflow: -4294964789 nr_pages=4294967295
WARNING: CPU: 2 PID: 3785 at mm/page_counter.c:56 page_counter_cancel+0xcf/0xe0 mm/page_counter.c:56
Modules linked in:
CPU: 2 PID: 3785 Comm: kworker/2:6 Not tainted 5.16.0-rc1-syzkaller #0
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-2 04/01/2014
Workqueue: events mptcp_worker
RIP: 0010:page_counter_cancel+0xcf/0xe0 mm/page_counter.c:56
Code: c7 04 24 00 00 00 00 45 31 f6 eb 97 e8 2a 2b b5 ff 4c 89 ea 48 89 ee 48 c7 c7 00 9e b8 89 c6 05 a0 c1 ba 0b 01 e8 95 e4 4b 07 <0f> 0b eb a8 4c 89 e7 e8 25 5a fb ff eb c7 0f 1f 00 41 56 41 55 49
RSP: 0018:ffffc90002d4f918 EFLAGS: 00010082
RAX: 0000000000000000 RBX: ffff88806a494120 RCX: 0000000000000000
RDX: ffff8880688c41c0 RSI: ffffffff815e8f28 RDI: fffff520005a9f15
RBP: ffffffff000009cb R08: 0000000000000000 R09: 0000000000000000
R10: ffffffff815e2cfe R11: 0000000000000000 R12: ffff88806a494120
R13: 00000000ffffffff R14: 0000000000000000 R15: 0000000000000001
FS: 0000000000000000(0000) GS:ffff88802cc00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000001b2de21000 CR3: 000000005ad59000 CR4: 0000000000150ee0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
page_counter_uncharge+0x2e/0x60 mm/page_counter.c:160
drain_stock+0xc1/0x180 mm/memcontrol.c:2219
refill_stock+0x139/0x2f0 mm/memcontrol.c:2271
__sk_mem_reduce_allocated+0x24d/0x550 net/core/sock.c:2945
__mptcp_rmem_reclaim net/mptcp/protocol.c:167 [inline]
__mptcp_mem_reclaim_partial+0x124/0x410 net/mptcp/protocol.c:975
mptcp_mem_reclaim_partial net/mptcp/protocol.c:982 [inline]
mptcp_alloc_tx_skb net/mptcp/protocol.c:1212 [inline]
mptcp_sendmsg_frag+0x18c6/0x2190 net/mptcp/protocol.c:1279
__mptcp_push_pending+0x232/0x720 net/mptcp/protocol.c:1545
mptcp_release_cb+0xfe/0x200 net/mptcp/protocol.c:2975
release_sock+0xb4/0x1b0 net/core/sock.c:3306
mptcp_worker+0x51e/0xc10 net/mptcp/protocol.c:2443
process_one_work+0x9b2/0x1690 kernel/workqueue.c:2298
worker_thread+0x658/0x11f0 kernel/workqueue.c:2445
kthread+0x405/0x4f0 kernel/kthread.c:327
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
</TASK>
__mptcp_mem_reclaim_partial() could call __mptcp_rmem_reclaim() with a
negative value, which passed that negative value to
__sk_mem_reduce_allocated() and triggered the splat above.
Check for a reclaim amount that is positive and large enough for
__mptcp_rmem_reclaim() to actually adjust rmem_fwd_alloc (much like
the sk_mem_reclaim_partial() code the function is based on).
v2: Use '>' instead of '>=', since SK_MEM_QUANTUM - 1 would get
right-shifted into nothing by __mptcp_rmem_reclaim.
Fixes: 6511882cdd82 ("mptcp: allocate fwd memory separately on the rx and tx path")
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/252
Reported-and-tested-by: syzbot+bc9e2d2dbcb347dd215a@syzkaller.appspotmail.com
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Michal Hocko <mhocko@suse.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
'ptr += 1;' was omitted in the original code.
If the DSS is the last option -- which is what we have most of the
time -- that's not an issue. But it is if we need to send something else
after like a RM_ADDR or an MP_PRIO.
Fixes: 1bff1e43a30e ("mptcp: optimize out option generation")
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
When these two options had to be sent -- which is not common -- the DSS
size was not being taken into account in the remaining size.
Additionally in this situation, the reported size was only the one of
the MP_FAIL which can cause issue if at the end, we need to write more
in the TCP options than previously said.
Here we use a dedicated variable for MP_FAIL size to keep the
WARN_ON_ONCE() just after.
Fixes: c25aeb4e0953 ("mptcp: MP_FAIL suboption sending")
Acked-and-tested-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Currently the msk->flags bitmask carries both state for the
mptcp_release_cb() - mostly touched under the mptcp data lock
- and others state info touched even outside such lock scope.
As a consequence, msk->flags is always manipulated with
atomic operations.
This change splits such bitmask in two separate fields, so
that we use plain bit operations when touching the
cb-related info.
The MPTCP_PUSH_PENDING bit needs additional care, as it is the
only CB related field currently accessed either under the mptcp
data lock or the mptcp socket lock.
Let's add another mask just for such bit's sake.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
We can simplify the join list handling leveraging the
mptcp_release_cb(): if we can acquire the msk socket
lock at mptcp_finish_join time, move the new subflow
directly into the conn_list, otherwise place it on join_list and
let the release_cb process such list.
Since pending MPJ connection are now always processed
in a timely way, we can avoid flushing the join list
every time we have to process all the current subflows.
Additionally we can now use the mptcp data lock to protect
the join_list, removing the additional spin lock.
Finally, the MPJ handshake is now always finalized under the
msk socket lock, we can drop the additional synchronization
between mptcp_finish_join() and mptcp_close().
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
If the MPTCP configuration allows for multiple subflows
creation, and the first additional subflows never reach
the fully established status - e.g. due to packets drop or
reset - the in kernel path manager do not move to the
next subflow.
This patch introduces a new PM helper to cope with MPJ
subflow creation failure and delay and hook it where appropriate.
Such helper triggers additional subflow creation, as needed
and updates the PM subflow counter, if the current one is
closing.
Additionally start all the needed additional subflows
as soon as the MPTCP socket is fully established, so we don't
have to cope with slow MPJ handshake blocking the next subflow
creation.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Include into the path manager status a bitmap tracking the list
of local endpoints still available - not yet used - for the
relevant mptcp socket.
Keep such map updated at endpoint creation/deletion time, so
that we can easily skip already used endpoint at local address
selection time.
The endpoint used by the initial subflow is lazyly accounted at
subflow creation time: the usage bitmap is be up2date before
endpoint selection and we avoid such unneeded task in some relevant
scenarios - e.g. busy servers accepting incoming subflows but
not creating any additional ones nor annuncing additional addresses.
Overall this allows for fair local endpoints usage in case of
subflow failure.
As a side effect, this patch also enforces that each endpoint
is used at most once for each mptcp connection.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Check for all MPJ variant at once, this reduces the number
of conditionals traversed on average and will simplify the
next patch.
No functional change intended.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Since full-mesh endpoint support, the reception of a single ADD_ADDR
option can cause multiple subflows creation. When such option is
accepted we increment 'add_addr_accepted' by one. When we received
a paired RM_ADDR option, we deleted all the relevant subflows,
decrementing 'add_addr_accepted' by one for each of them.
We have a similar issue for 'local_addr_used'
Fix them moving the pm endpoint accounting outside the subflow
traversal.
Fixes: 1a0d6136c5f0 ("mptcp: local addresses fullmesh")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Handle explicitly AF_UNSPEC in mptcp_stream_connnect() to
allow user-space to disconnect established MPTCP connections
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
After the previous patch, msk->subflow will never be deleted during
the whole msk lifetime. We don't need anymore to acquire references to
it in mptcp_stream_accept() and we can use the listener subflow accept
queue to simplify mptcp_poll() for listener socket.
Overall this removes a lock pair and 4 more atomic operations per
accept().
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The current mptcp_disconnect() implementation lacks several
steps, we additionally need to reset the msk socket state
and flush the subflow list.
Factor out the needed helper to avoid code duplication.
Additionally ensure that the initial subflow is disposed
only after mptcp_close(), just reset it at disconnect time.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Allow the MPTCP xmit path to add MP_FASTCLOSE suboption
on RST egress packets.
Additionally reorder related options writing to reduce
the number of conditionals required in the fast path.
Co-developed-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
After shutdown, for fallback MPTCP sockets, we always have
write_seq == snd_una+1
The above will foul OUTQ ioctl(). Keep snd_una in sync with
write_seq even after shutdown.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
entry->addr.id is u8 with a range from 0 to 255 and MAX_ADDR_ID is 255.
We should drop both false expressions of (entry->addr.id > MAX_ADDR_ID).
We should also remove the obsolete parentheses in the first if branch.
Use U8_MAX for MAX_ADDR_ID and add a comment to show the link to
mptcp_addr_info.id as suggested by Mr. Matthieu Baerts.
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Jean Sacren <sakiwit@gmail.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The MPTCP packet scheduler has sub-optimal behavior with asymmetric
subflows: if the faster subflow-level cwin is closed, the packet
scheduler can enqueue "too much" data on a slower subflow.
When all the data on the faster subflow is acked, if the mptcp-level
cwin is closed, and link utilization becomes suboptimal.
The solution is implementing blest-like[1] HoL-blocking estimation,
transmitting only on the subflow with the shorter estimated time to
flush the queued memory. If such subflows cwin is closed, we wait
even if other subflows are available.
This is quite simpler than the original blest implementation, as we
leverage the pacing rate provided by the TCP socket. To get a more
accurate estimation for the subflow linger-time, we maintain a
per-subflow weighted average of such info.
Additionally drop magic numbers usage in favor of newly defined
macros and use more meaningful names for status variable.
[1] http://dl.ifip.org/db/conf/networking/networking2016/1570234725.pdf
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/137
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
No conflicts.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
__mptcp_push_pending() may call mptcp_flush_join_list() with subflow
socket lock held. If such call hits mptcp_sockopt_sync_all() then
subsequently __mptcp_sockopt_sync() could try to lock the subflow
socket for itself, causing a deadlock.
sysrq: Show Blocked State
task:ss-server state:D stack: 0 pid: 938 ppid: 1 flags:0x00000000
Call Trace:
<TASK>
__schedule+0x2d6/0x10c0
? __mod_memcg_state+0x4d/0x70
? csum_partial+0xd/0x20
? _raw_spin_lock_irqsave+0x26/0x50
schedule+0x4e/0xc0
__lock_sock+0x69/0x90
? do_wait_intr_irq+0xa0/0xa0
__lock_sock_fast+0x35/0x50
mptcp_sockopt_sync_all+0x38/0xc0
__mptcp_push_pending+0x105/0x200
mptcp_sendmsg+0x466/0x490
sock_sendmsg+0x57/0x60
__sys_sendto+0xf0/0x160
? do_wait_intr_irq+0xa0/0xa0
? fpregs_restore_userregs+0x12/0xd0
__x64_sys_sendto+0x20/0x30
do_syscall_64+0x38/0x90
entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x7f9ba546c2d0
RSP: 002b:00007ffdc3b762d8 EFLAGS: 00000246 ORIG_RAX: 000000000000002c
RAX: ffffffffffffffda RBX: 00007f9ba56c8060 RCX: 00007f9ba546c2d0
RDX: 000000000000077a RSI: 0000000000e5e180 RDI: 0000000000000234
RBP: 0000000000cc57f0 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00007f9ba56c8060
R13: 0000000000b6ba60 R14: 0000000000cc7840 R15: 41d8685b1d7901b8
</TASK>
Fix the issue by using __mptcp_flush_join_list() instead of plain
mptcp_flush_join_list() inside __mptcp_push_pending(), as suggested by
Florian. The sockopt sync will be deferred to the workqueue.
Fixes: 1b3e7ede1365 ("mptcp: setsockopt: handle SO_KEEPALIVE and SO_PRIORITY")
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/244
Suggested-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Maxim Galaganov <max@internet.ru>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The mptcp ULP extension relies on sk->sk_sock_kern being set correctly:
It prevents setsockopt(fd, IPPROTO_TCP, TCP_ULP, "mptcp", 6); from
working for plain tcp sockets (any userspace-exposed socket).
But in case of fallback, accept() can return a plain tcp sk.
In such case, sk is still tagged as 'kernel' and setsockopt will work.
This will crash the kernel, The subflow extension has a NULL ctx->conn
mptcp socket:
BUG: KASAN: null-ptr-deref in subflow_data_ready+0x181/0x2b0
Call Trace:
tcp_data_ready+0xf8/0x370
[..]
Fixes: cf7da0d66cc1 ("mptcp: Create SUBFLOW socket for incoming connections")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
TCP_ULP setsockopt cannot be used for mptcp because its already
used internally to plumb subflow (tcp) sockets to the mptcp layer.
syzbot managed to trigger a crash for mptcp connections that are
in fallback mode:
KASAN: null-ptr-deref in range [0x0000000000000020-0x0000000000000027]
CPU: 1 PID: 1083 Comm: syz-executor.3 Not tainted 5.16.0-rc2-syzkaller #0
RIP: 0010:tls_build_proto net/tls/tls_main.c:776 [inline]
[..]
__tcp_set_ulp net/ipv4/tcp_ulp.c:139 [inline]
tcp_set_ulp+0x428/0x4c0 net/ipv4/tcp_ulp.c:160
do_tcp_setsockopt+0x455/0x37c0 net/ipv4/tcp.c:3391
mptcp_setsockopt+0x1b47/0x2400 net/mptcp/sockopt.c:638
Remove support for TCP_ULP setsockopt.
Fixes: d9e4c1291810 ("mptcp: only admit explicitly supported sockopt")
Reported-by: syzbot+1fd9b69cde42967d1add@syzkaller.appspotmail.com
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
MPTCP can change sk_net_refcnt after sock_create_kern() call.
We need to change its corresponding get_net() to avoid
a splat at release time, as in :
refcount_t: decrement hit 0; leaking memory.
WARNING: CPU: 0 PID: 3599 at lib/refcount.c:31 refcount_warn_saturate+0xbf/0x1e0 lib/refcount.c:31
Modules linked in:
CPU: 1 PID: 3599 Comm: syz-fuzzer Not tainted 5.16.0-rc4-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:refcount_warn_saturate+0xbf/0x1e0 lib/refcount.c:31
Code: 1d b1 99 a1 09 31 ff 89 de e8 5d 3a 9c fd 84 db 75 e0 e8 74 36 9c fd 48 c7 c7 60 00 05 8a c6 05 91 99 a1 09 01 e8 cc 4b 27 05 <0f> 0b eb c4 e8 58 36 9c fd 0f b6 1d 80 99 a1 09 31 ff 89 de e8 28
RSP: 0018:ffffc90001f5fab0 EFLAGS: 00010286
RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
RDX: ffff888021873a00 RSI: ffffffff815f1e28 RDI: fffff520003ebf48
RBP: 0000000000000004 R08: 0000000000000000 R09: 0000000000000000
R10: ffffffff815ebbce R11: 0000000000000000 R12: 1ffff920003ebf5b
R13: 00000000ffffffef R14: ffffffff8d2fcd94 R15: ffffc90001f5fd10
FS: 000000c00008a090(0000) GS:ffff8880b9d00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f0a5b59e300 CR3: 000000001cbe6000 CR4: 00000000003506e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
__refcount_dec include/linux/refcount.h:344 [inline]
refcount_dec include/linux/refcount.h:359 [inline]
ref_tracker_free+0x4fe/0x610 lib/ref_tracker.c:101
netns_tracker_free include/net/net_namespace.h:327 [inline]
put_net_track include/net/net_namespace.h:341 [inline]
__sk_destruct+0x4a6/0x920 net/core/sock.c:2042
sk_destruct+0xbd/0xe0 net/core/sock.c:2058
__sk_free+0xef/0x3d0 net/core/sock.c:2069
sk_free+0x78/0xa0 net/core/sock.c:2080
sock_put include/net/sock.h:1911 [inline]
__mptcp_close_ssk+0x435/0x590 net/mptcp/protocol.c:2276
__mptcp_destroy_sock+0x35f/0x830 net/mptcp/protocol.c:2702
mptcp_close+0x5f8/0x7f0 net/mptcp/protocol.c:2750
inet_release+0x12e/0x280 net/ipv4/af_inet.c:428
inet6_release+0x4c/0x70 net/ipv6/af_inet6.c:476
__sock_release+0xcd/0x280 net/socket.c:649
sock_close+0x18/0x20 net/socket.c:1314
__fput+0x286/0x9f0 fs/file_table.c:280
task_work_run+0xdd/0x1a0 kernel/task_work.c:164
tracehook_notify_resume include/linux/tracehook.h:189 [inline]
exit_to_user_mode_loop kernel/entry/common.c:175 [inline]
exit_to_user_mode_prepare+0x27e/0x290 kernel/entry/common.c:207
__syscall_exit_to_user_mode_work kernel/entry/common.c:289 [inline]
syscall_exit_to_user_mode+0x19/0x60 kernel/entry/common.c:300
do_syscall_64+0x42/0xb0 arch/x86/entry/common.c:86
entry_SYSCALL_64_after_hwframe+0x44/0xae
Fixes: ffa84b5ffb37 ("net: add netns refcount tracker to struct sock")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Cc: Mat Martineau <mathew.j.martineau@linux.intel.com>
Cc: Florian Westphal <fw@strlen.de>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Link: https://lore.kernel.org/r/20211214043208.3543046-1-eric.dumazet@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Currently, when deleting an endpoint the netlink PM treverses
all the local MPTCP sockets, regardless of their status.
If an MPTCP listener socket is bound to the IP matching the
delete endpoint, the listener TCP socket will be closed.
That is unexpected, the PM should only affect data subflows.
Additionally, syzbot was able to trigger a NULL ptr dereference
due to the above:
general protection fault, probably for non-canonical address 0xdffffc0000000003: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000018-0x000000000000001f]
CPU: 1 PID: 6550 Comm: syz-executor122 Not tainted 5.16.0-rc4-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:__lock_acquire+0xd7d/0x54a0 kernel/locking/lockdep.c:4897
Code: 0f 0e 41 be 01 00 00 00 0f 86 c8 00 00 00 89 05 69 cc 0f 0e e9 bd 00 00 00 48 b8 00 00 00 00 00 fc ff df 48 89 da 48 c1 ea 03 <80> 3c 02 00 0f 85 f3 2f 00 00 48 81 3b 20 75 17 8f 0f 84 52 f3 ff
RSP: 0018:ffffc90001f2f818 EFLAGS: 00010016
RAX: dffffc0000000000 RBX: 0000000000000018 RCX: 0000000000000000
RDX: 0000000000000003 RSI: 0000000000000000 RDI: 0000000000000001
RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000001
R10: 0000000000000000 R11: 000000000000000a R12: 0000000000000000
R13: ffff88801b98d700 R14: 0000000000000000 R15: 0000000000000001
FS: 00007f177cd3d700(0000) GS:ffff8880b9d00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f177cd1b268 CR3: 000000001dd55000 CR4: 0000000000350ee0
Call Trace:
<TASK>
lock_acquire kernel/locking/lockdep.c:5637 [inline]
lock_acquire+0x1ab/0x510 kernel/locking/lockdep.c:5602
__raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
_raw_spin_lock_irqsave+0x39/0x50 kernel/locking/spinlock.c:162
finish_wait+0xc0/0x270 kernel/sched/wait.c:400
inet_csk_wait_for_connect net/ipv4/inet_connection_sock.c:464 [inline]
inet_csk_accept+0x7de/0x9d0 net/ipv4/inet_connection_sock.c:497
mptcp_accept+0xe5/0x500 net/mptcp/protocol.c:2865
inet_accept+0xe4/0x7b0 net/ipv4/af_inet.c:739
mptcp_stream_accept+0x2e7/0x10e0 net/mptcp/protocol.c:3345
do_accept+0x382/0x510 net/socket.c:1773
__sys_accept4_file+0x7e/0xe0 net/socket.c:1816
__sys_accept4+0xb0/0x100 net/socket.c:1846
__do_sys_accept net/socket.c:1864 [inline]
__se_sys_accept net/socket.c:1861 [inline]
__x64_sys_accept+0x71/0xb0 net/socket.c:1861
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x7f177cd8b8e9
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 b1 14 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:00007f177cd3d308 EFLAGS: 00000246 ORIG_RAX: 000000000000002b
RAX: ffffffffffffffda RBX: 00007f177ce13408 RCX: 00007f177cd8b8e9
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000003
RBP: 00007f177ce13400 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00007f177ce1340c
R13: 00007f177cde1004 R14: 6d705f706374706d R15: 0000000000022000
</TASK>
Fix the issue explicitly skipping MPTCP socket in TCP_LISTEN
status.
Reported-and-tested-by: syzbot+e4d843bb96a9431e6331@syzkaller.appspotmail.com
Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Fixes: 740d798e8767 ("mptcp: remove id 0 address")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Link: https://lore.kernel.org/r/ebc7594cdd420d241fb2172ddb8542ba64717657.1639238695.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
First, add cork and nodelay fields to the mptcp_sock structure
so they can be used in sync_socket_options(), and fill them on setsockopt
while holding the msk socket lock.
Then, on setsockopt set proper tcp_sk(ssk)->nonagle values for subflows
by calling __tcp_sock_set_cork() or __tcp_sock_set_nodelay() on the ssk
while holding the ssk socket lock.
tcp_push_pending_frames() will be invoked on the ssk if a cork was cleared
or nodelay was set. Also set MPTCP_PUSH_PENDING bit by calling
mptcp_check_and_set_pending(). This will lead to __mptcp_push_pending()
being called inside mptcp_release_cb() with new tcp_sk(ssk)->nonagle.
Also add getsockopt support for TCP_CORK and TCP_NODELAY.
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Maxim Galaganov <max@internet.ru>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Expose the mptcp_check_and_set_pending() function for use inside MPTCP
sockopt code. The next patch will call it when TCP_CORK is cleared or
TCP_NODELAY is set on the MPTCP socket in order to push pending data
from mptcp_release_cb().
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Maxim Galaganov <max@internet.ru>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
earlier patch added IP_TOS setsockopt support, this allows to get
the value set by earlier setsockopt.
Extends mptcp_put_int_option to handle u8 input/output by
adding required cast.
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
a non-zero 'id' is sufficient to identify MPTCP endpoints: allow changing
the value of 'backup' bit by simply specifying the endpoint id.
Link: https://github.com/multipath-tcp/mptcp_net-next/issues/158
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Allows to query in-sequence data ready for read(), total bytes in
write queue and total bytes in write queue that have not yet been sent.
v2: remove unneeded READ_ONCE() (Paolo Abeni)
v3: check for new data unconditionally in SIOCINQ ioctl (Mat Martineau)
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Support the TCP_INQ setsockopt.
This is a boolean that tells recvmsg path to include the remaining
in-sequence bytes in the cmsg data.
v2: do not use CB(skb)->offset, increment map_seq instead (Paolo Abeni)
v3: adjust CB(skb)->map_seq when taking skb from ofo queue (Paolo Abeni)
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/224
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
drivers/net/ipa/ipa_main.c
8afc7e471ad3 ("net: ipa: separate disabling setup from modem stop")
76b5fbcd6b47 ("net: ipa: kill ipa_modem_init()")
Duplicated include, drop one.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Scheduling a delack in mptcp_established_options_mp() is
not a good idea: such function is called by tcp_send_ack() and
the pending delayed ack will be cleared shortly after by the
tcp_event_ack_sent() call in __tcp_transmit_skb().
Instead use the mptcp delegated action infrastructure to
schedule the delayed ack after the current bh processing completes.
Additionally moves the schedule_3rdack_retransmission() helper
into protocol.c to avoid making it visible in a different compilation
unit.
Fixes: ec3edaa7ca6ce02f ("mptcp: Add handling of outgoing MP_JOIN requests")
Reviewed-by: Mat Martineau <mathew.j.martineau>@linux.intel.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
To compute the rtx timeout schedule_3rdack_retransmission() does multiple
things in the wrong way: srtt_us is measured in usec/8 and the timeout
itself is an absolute value.
Fixes: ec3edaa7ca6ce02f ("mptcp: Add handling of outgoing MP_JOIN requests")
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Mat Martineau <mathew.j.martineau>@linux.intel.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
These options also need to be set before bind, so do the sync of msk to
new ssk socket a bit earlier.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
SOL_IP provides a way to configure network layer attributes in a
socket. This patch adds support for IP_TOS for setsockopt(.. ,SOL_IP, ..)
Support for SOL_IP is added in mptcp_setsockopt() and IP_TOS is handled in
a private function. The idea here is to take in the value passed for IP_TOS
and set it to the current subflow, open subflows as well new subflows that
might be created after the initial call to setsockopt(). This sync is done
using sync_socket_options(.., ssk) and setting the value of tos using
__ip_sock_set_tos(ssk,..).
The patch has been tested using the packetdrill script here -
https://github.com/multipath-tcp/mptcp_net-next/issues/220#issuecomment-947863717
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/220
Signed-off-by: Poorva Sonparote <psonparo@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
MPTCP hard codes it, let us instead provide this helper.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
tcp_memory_allocated and tcp_sockets_allocated often share
a common cache line, source of false sharing.
Also take care of udp_memory_allocated and mptcp_sockets_allocated.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
include/net/sock.h
7b50ecfcc6cd ("net: Rename ->stream_memory_read to ->sock_is_readable")
4c1e34c0dbff ("vsock: Enable y2038 safe timeval for timeout")
drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c
0daa55d033b0 ("octeontx2-af: cn10k: debugfs for dumping LMTST map table")
e77bcdd1f639 ("octeontx2-af: Display all enabled PF VF rsrc_alloc entries.")
Adjacent code addition in both cases, keep both.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
using packetdrill it's possible to observe that the receiver key contains
random values when clients transmit MP_CAPABLE with data and checksum (as
specified in RFC8684 ยง3.1). Fix the layout of mptcp_out_options, to avoid
using the skb extension copy when writing the MP_CAPABLE sub-option.
Fixes: d7b269083786 ("mptcp: shrink mptcp_out_options struct")
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/233
Reported-by: Poorva Sonparote <psonparo@redhat.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Link: https://lore.kernel.org/r/20211027203855.264600-1-mathew.j.martineau@linux.intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Setting skb->ip_summed to CHECKSUM_PARTIAL can be centralized
in tcp_stream_alloc_skb() and __mptcp_do_alloc_tx_skb()
instead of being done multiple times.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|