From: Matthieu Baerts <matttbe@kernel.org>
To: MPTCP Linux <mptcp@lists.linux.dev>
Subject: [Weekly meetings] MoM - 5th of November 2025
Date: Wed, 12 Nov 2025 18:58:18 +0100 [thread overview]
Message-ID: <5e7430ff-ef27-49ef-a295-875a8c6fdcd1@kernel.org> (raw)
Hello everyone,
Last week, we had our new meeting with Mat (Cruise), Paolo (Redhat) and
myself (NGI0 Core).
Thanks again for this new good meeting!
Here are the minutes of the meeting:
Accepted patches:
- The list of accepted patches can be seen on PatchWork:
https://patchwork.kernel.org/project/mptcp/list/?state=3
netdev (if mptcp ML is in cc) (by: Matthieu Baerts):
14287659: [net,1/4] mptcp: drop bogus optimization in __mptcp_check_push()
14287660: [net,2/4] mptcp: fix MSG_PEEK stream corruption
14287661: [net,3/4] mptcp: restore window probe
14287662: [net,4/4] mptcp: zero window probe mib:
- Series: mptcp: various rare sending issues
- Submitter: Matthieu Baerts
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=1016623&state=*&archive=both
14287911: [net,v3,1/4] mptcp: fix subflow rcvbuf adjust
14287912: [net,v3,2/4] trace: tcp: add three metrics to
trace_tcp_rcvbuf_grow()
14287913: [net,v3,3/4] tcp: add newval parameter to tcp_rcvbuf_grow()
14287914: [net,v3,4/4] tcp: fix too slow tcp_rcvbuf_grow() action:
- Series: tcp: fix receive autotune again
- Submitter: Matthieu Baerts
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=1016697&state=*&archive=both
14295026: [net-next,1/4] mptcp: pm: in-kernel: record fullmesh endp nb
14295027: [net-next,2/4] mptcp: pm: in kernel: only use fullmesh endp if any
14295028: [net-next,3/4] selftests: mptcp: join: do_transfer: reduce
code dup
14295029: [net-next,4/4] selftests: mptcp: join: validate extra bind cases:
- Series: mptcp: pm: in-kernel: fullmesh endp nb + bind cases
- Submitter: Matthieu Baerts
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=1018552&state=*&archive=both
our repo (by: GangYan, Matthieu Baerts, Paolo Abeni):
14278733: [v2,mptcp-net,1/2] mptcp: restore window probe
14278734: [v2,mptcp-net,2/2] mptcp: zero window probe mib:
- Series: mptcp: restore zero window probe
- Submitter: Paolo Abeni
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=1014382&state=*&archive=both
14279404: [v6,mptcp-next,01/11] mptcp: drop bogus optimization in
__mptcp_check_push()
14279409: [v6,mptcp-next,03/11] mptcp: cleanup fallback data fin reception
14279407: [v6,mptcp-next,04/11] mptcp: cleanup fallback dummy mapping
generation
14279406: [v6,mptcp-next,05/11] mptcp: fix MSG_PEEK stream corruption
14279408: [v6,mptcp-next,06/11] mptcp: ensure the kernel PM does not
take action too late
14279411: [v6,mptcp-next,07/11] mptcp: do not miss early first subflow
close event notification.
14279410: [v6,mptcp-next,08/11] mptcp: make mptcp_destroy_common() static
14279412: [v6,mptcp-next,09/11] mptcp: drop the __mptcp_data_ready() helper
14283178: [mptcp-next] Squash to "selftests: mptcp: join: validate extra
bind cases":
- Submitter: Matthieu Baerts
- URL:
https://patchwork.kernel.org/project/mptcp/patch/20251024-sft-fix-bind-timeout-v1-1-cb81a67b222c@kernel.org/
14295242: [mptcp-net,v2,1/6] selftests: mptcp: connect: fix fallback
note due to OoO
14295243: [mptcp-net,v2,2/6] selftests: mptcp: join: rm: set backup flag
14295244: [mptcp-net,v2,3/6] selftests: mptcp: join: endpoints: longer
transfer
14295245: [mptcp-net,v2,4/6] selftests: mptcp: join: userspace: longer
transfer
14295246: [mptcp-net,v2,5/6] selftests: mptcp: join: fastclose: drop
plain RST
14295247: [mptcp-net,v2,6/6] selftests: mptcp: connect: trunc: read all
recv data:
- Series: selftests: mptcp: join: fix flaky tests
- Submitter: Matthieu Baerts
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=1018612&state=*&archive=both
14297102: [mptcp-next,1/5] selftests: mptcp: lib: introduce
'nstat_{init,get}'
14297103: [mptcp-next,2/5] selftests: mptcp: lib: remove stats files args
14297104: [mptcp-next,3/5] selftests: mptcp: lib: stats: remove nstat
rate columns
14297105: [mptcp-next,4/5] selftests: mptcp: join: dump stats from history
14297106: [mptcp-next,5/5] selftests: mptcp: lib: get counters from
nstat history:
- Series: selftests: mptcp: get counters from nstat history
- Submitter: Matthieu Baerts
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=1019082&state=*&archive=both
14298228: [mptcp-net] mptcp: fix address removal logic in
mptcp_pm_nl_rm_addr:
- Submitter: GangYan
- URL:
https://patchwork.kernel.org/project/mptcp/patch/20251104123416.395362-1-gang.yan@linux.dev/
14298615: [mptcp-next,v8,1/4] mptcp: handle first subflow closing
consistently
14298614: [mptcp-next,v8,2/4] mptcp: borrow forward memory from subflow
14298616: [mptcp-next,v8,3/4] mptcp: introduce mptcp-level backlog
14298617: [mptcp-next,v8,4/4] mptcp: leverage the backlog for RX packet
processing:
- Series: mptcp: introduce backlog processing
- Submitter: Paolo Abeni
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=1019535&state=*&archive=both
Pending patches:
- The list of pending patches can be seen on PatchWork:
https://patchwork.kernel.org/project/mptcp/list/?state=*
netdev (if mptcp ML is in cc) (by: Jiayuan Chen):
14299873: [net,v4,1/3] mptcp: disallow MPTCP subflows from sockmap
14299851: [net,v4,2/3] net,mptcp: fix proto fallback detection with BPF
14299874: [net,v4,3/3] selftests/bpf: Add mptcp test with sockmap:
- Series: mptcp: Fix conflicts between MPTCP and sockmap
- Submitter: Jiayuan Chen
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=1019848&state=*&archive=both
14296835: [6.12.y-5.10.y] selftests: mptcp: connect modes: re-add exec mode:
- State: New
- Submitter: Matthieu Baerts
- URL:
https://patchwork.kernel.org/project/mptcp/patch/20251103165433.6396-2-matttbe@kernel.org/
- cannot be applied because they use Quilt
14298196: [6.12.y] mptcp: fix MSG_PEEK stream corruption:
- State: New
- Submitter: Matthieu Baerts
- URL:
https://patchwork.kernel.org/project/mptcp/patch/20251104121515.1093006-2-matttbe@kernel.org/
our repo (by: Christoph Paasch, Gang Yan, Geliang Tang, Gregory
Detal, Matthieu Baerts, Paolo Abeni):
12282223: [RESEND,RFC,mptpcp-next] mptcp: add ooo prune support:
13070467: [mptcp-next,v4,1/2] mptcp: update mptcp_push_release
13070468: [mptcp-next,v4,2/2] mptcp: retrans for redundant sends:
- State: Changes requested
- Series: BPF redundant scheduler, part 3
- Submitter: Geliang Tang
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=703615&state=*&archive=both
- status: same as last time
13305089: [mptcp-next,1/6] mptcp: sockopt: move tcp_inq code to a
dedicated function
13305091: [mptcp-next,2/6] mptcp: sockopt: update supported list
13305093: [mptcp-next,3/6] mptcp: sockopt: get val in a generic way
13305090: [mptcp-next,4/6] mptcp: sockopt: add missing getsockopt() options
13305094: [mptcp-next,5/6] mptcp: sockopt: set val in a generic way
13305092: [mptcp-next,6/6] mptcp: sockopt: support IP_TTL &
IPV6_UNICAST_HOPS:
- State: Changes requested
- Series: mptcp: sockopt: uniform code to get/set values
- Submitter: Matthieu Baerts
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=763505&state=*&archive=both
- status: same as last time
13583230: [RFC] mptcp: react scheduler when subflow events pop up:
- State: Changes requested
- Submitter: Geliang Tang
- URL:
https://patchwork.kernel.org/project/mptcp/patch/433320c3a9db77bea53a34fc9c43a3c7e3320399.1709693691.git.tanggeliang@kylinos.cn/
- status: same as last time
13675217: [mptcp-next,RFC,1/4] mptcp: add push sched callback
13675218: [mptcp-next,RFC,2/4] mptcp: use new push callback to schedule
chunks
13675219: [mptcp-next,RFC,3/4] mptcp: bpf: allow to write to
mptcp_sched_chunk
13675220: [mptcp-next,RFC,4/4] selftests/bpf: mptcp RR: send 1 MSS on
each subflow:
- State: RFC
- Series: mptcp: update scheduler API
- Submitter: Gregory Detal
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=856158&state=*&archive=both
- status: same as last time
13932555: [1/3] mptcp: fix invalid addr occupy 'add_addr_accepted'
13932556: [2/3] selftests: mptcp: relax conditions in SYN_TX validation
13932546: [3/3] selftests: mptcp: add a testcase for invalid addr timeout:
- State: Changes requested
- Series: [1/3] mptcp: fix invalid addr occupy 'add_addr_accepted'
- Submitter: Gang Yan
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=923771&state=*&archive=both
- status: same as last time
14024767: [mptcp-next,v1,1/4] bpf: Add mptcp path manager struct_ops
14024768: [mptcp-next,v1,2/4] bpf: Export mptcp path manager kfuncs
14024769: [mptcp-next,v1,3/4] selftests/bpf: Add mptcp netlink pm subtest
14024770: [mptcp-next,v1,4/4] selftests/bpf: Add mptcp bpf_netlink pm
subtest:
- State: Changes requested
- Series: BPF path manager, part 7
- Submitter: Geliang Tang
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=946095&state=*&archive=both
- status: same as last time
14053397: [mptcp-next,v3,1/3] mptcp: pm: userspace: drop
delete_local_addr helper
14053398: [mptcp-next,v3,2/3] selftests: mptcp: join: update userspace
dump_addr outputs
14053399: [mptcp-next,v3,3/3] selftests: mptcp: userspace: send RM_ADDR
between server and client:
- State: Under review
- Series: only remove entry from local_addr_list when sending a
REMOVE_ADDR
- Submitter: Geliang Tang
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=953862&state=*&archive=both
- status: same as last time
14102233: [mptcp-next,v3,1/3] selftests/bpf: Add sndbuf for send_recv_data
14102235: [mptcp-next,v3,2/3] Squash to "selftests/bpf: Add bpf
scheduler test"
14102234: [mptcp-next,v3,3/3] DO-NOT-MERGE: selftests/bpf: Increase
total_bytes of bpf sched tests:
- State: Changes requested
- Series: send() fails with EAGAIN in blocking IO mode "487
- Submitter: Geliang Tang
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=967210&state=*&archive=both
- status: same as last time
14167004: [mptcp-next] mptcp: remove redundant orig_offset in
carve_data_frag:
- State: Changes requested
- Submitter: Geliang Tang
- URL:
https://patchwork.kernel.org/project/mptcp/patch/9c663fc37767d9036379afdbc83cd8874a7b2fc9.1753337130.git.tanggeliang@kylinos.cn/
- status: same as last time
14175678: [mptcp-next,1/2] mptcp: Add support for SO_MAX_PACING_RATE
14175679: [mptcp-next,2/2] selftests: Add tests for SO_MAX_PACING_RATE:
- State: Changes requested
- Series: mptcp: Add support for SO_MAX_PACING_RATE
- Submitter: Christoph Paasch
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=987909&state=*&archive=both
- status: same as last time
14185125: [mptcp-next,1/6] mptcp: add helper for setsockopt on first
subflow only
14185126: [mptcp-next,2/6] mptcp: handle SOL_SOCKET options on first
subflow only
14185128: [mptcp-next,3/6] mptcp: refactor IP socket option handling for
first subflow
14185127: [mptcp-next,4/6] mptcp: extract IPv6 socket option handling
for first subflow
14185129: [mptcp-next,5/6] mptcp: handle SOL_TCP options on all subflows
14185130: [mptcp-next,6/6] mptcp: handle IP socket options on all subflows:
- State: Needs ack
- Series: refactor MPTCP socket option handling
- Submitter: Geliang Tang
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=990414&state=*&archive=both
- status: same as last time
14185142: [mptcp-next,v2,1/3] mptcp: add IPV6_TCLASS socket option support
14185143: [mptcp-next,v2,2/3] mptcp: add IP_TTL socket option support
14185144: [mptcp-next,v2,3/3] mptcp: add IPV6_UNICAST_HOPS socket option
support:
- State: Needs ack
- Series: mptcp: add more sockopts
- Submitter: Geliang Tang
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=990418&state=*&archive=both
- status: same as last time
14197780: [mptcp-next,v2,01/10] selftests: mptcp: sockopt: improve test
output clarity
14197781: [mptcp-next,v2,02/10] selftests: mptcp: sockopt: add
SO_REUSEADDR test helper
14197782: [mptcp-next,v2,03/10] selftests: mptcp: sockopt: add
SO_REUSEPORT test
14197783: [mptcp-next,v2,04/10] selftests: mptcp: sockopt: add
SO_BINDTODEVICE test
14197784: [mptcp-next,v2,05/10] selftests: mptcp: sockopt: add
SO_BINDTOIFINDEX test
14197785: [mptcp-next,v2,06/10] selftests: mptcp: sockopt: add
IP_FREEBIND tests
14197786: [mptcp-next,v2,07/10] selftests: mptcp: sockopt: add
IP_TRANSPARENT tests
14197787: [mptcp-next,v2,08/10] selftests: mptcp: sockopt: add
IP_BIND_ADDRESS_NO_PORT test
14197788: [mptcp-next,v2,09/10] selftests: mptcp: sockopt: add
IP_LOCAL_PORT_RANGE test
14197789: [mptcp-next,v2,10/10] selftests: mptcp: sockopt: add
IPV6_V6ONLY test:
- State: Changes requested
- Series: selftests: cover more MPTCP socket options
- Submitter: Geliang Tang
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=993905&state=*&archive=both
- status: same as last time
14215849: [mptcp-net,v2,1/3] selftests: mptcp: close server file descriptor
14215850: [mptcp-net,v2,2/3] selftests: mptcp: close IPC descriptor on
server side
14229120: [mptcp-next,v4,1/9] selftests: mptcp: sockopt: replace
/dev/urandom with getrandom
14229121: [mptcp-next,v4,2/9] selftests: mptcp: sockopt: add protocol
arguments
14229122: [mptcp-next,v4,3/9] selftests: mptcp: sockopt: add inq argument
14229123: [mptcp-next,v4,4/9] selftests: mptcp: sockopt: rename pipefd
to ipcfd
14229124: [mptcp-next,v4,5/9] selftests: mptcp: sockopt: set TCP_INQ sockopt
14229125: [mptcp-next,v4,6/9] selftests: mptcp: sockopt: handle INQ on
server side
14229126: [mptcp-next,v4,7/9] selftests: mptcp: sockopt: handle INQ on
client side
14229127: [mptcp-next,v4,8/9] selftests: mptcp: sockopt: replace
mptcp_inq with sockopt
14229128: [mptcp-next,v4,9/9] selftests: mptcp: sockopt: remove obsolete
mptcp_inq:
- State: Changes requested
- Series: selftests: consolidate TCP_INQ testing into sockopt
- Submitter: Geliang Tang
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=1001827&state=*&archive=both
- status: same as last time
14280548: [mptcp-next,v13,1/8] mptcp: add eat_recv_skb helper
14280549: [mptcp-next,v13,2/8] mptcp: implement .read_sock
14280550: [mptcp-next,v13,3/8] tcp: add recv_should_stop helper
14280551: [mptcp-next,v13,4/8] mptcp: use recv_should_stop helper
14280552: [mptcp-next,v13,5/8] tcp: export tcp_splice_state
14280553: [mptcp-next,v13,6/8] mptcp: implement .splice_read
14280554: [mptcp-next,v13,7/8] selftests: mptcp: add splice io mode
14280555: [mptcp-next,v13,8/8] selftests: mptcp: connect: cover splice mode:
- State: New
- Series: implement mptcp read_sock
- Submitter: Geliang Tang
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=1014823&state=*&archive=both
- Mat will have a look at it
14299105: [v2,mptcp-next,1/7] trace: mptcp: add mptcp_rcvbuf_grow tracepoint
14299106: [v2,mptcp-next,2/7] mptcp: avoid unneeded subflow-level drops.
14299107: [v2,mptcp-next,3/7] mptcp: fix receive space timestamp
initialization.
14299108: [v2,mptcp-next,4/7] mptcp: consolidate rcv space init
14299109: [v2,mptcp-next,5/7] mptcp: better rcv space initialization
14299110: [v2,mptcp-next,6/7] mptcp: better mptcp-level rtt estimator
14299111: [v2,mptcp-next,7/7] mptcp: add receive queue awareness in
tcp_rcv_space_adjust():
- State: New
- Series: mptcp: autotune related improvement
- Submitter: Paolo Abeni
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=1019624&state=*&archive=both
14299674: [mptcp-next,v6,1/7] tcp: md5: remove redundant l3flag variable
14299675: [mptcp-next,v6,2/7] mptcp: handle TCP_MAXSEG getsockopt in
common case
14299676: [mptcp-next,v6,3/7] mptcp: setsockopt support for TCP_MD5SIG
14299677: [mptcp-next,v6,4/7] selftests: mptcp: sockopt: add md5 argument
14299678: [mptcp-next,v6,5/7] selftests: mptcp: sockopt: implement
TCP_MD5SIG support
14299679: [mptcp-next,v6,6/7] selftests: mptcp: sockopt: skip getsockopt
for MD5 tests
14299680: [mptcp-next,v6,7/7] selftests: mptcp: sockopt: add TCP_MD5SIG
test cases:
- State: New
- Series: mptcp: TCP_MD5SIG support
- Submitter: Geliang Tang
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=1019796&state=*&archive=both
- Matth: will check that
14300058: [mptcp-net,1/2] mptcp: decouple mptcp fastclose from tcp close
14300059: [mptcp-net,2/2] mptcp: fix duplicate reset on fastclose:
- State: New
- Series: [mptcp-net,1/2] mptcp: decouple mptcp fastclose from tcp close
- Submitter: Paolo Abeni
- URL:
https://patchwork.kernel.org/project/mptcp/list/?series=1019919&state=*&archive=both
Issues on GitHub:
https://github.com/multipath-tcp/mptcp_net-next/issues/
Recently opened (latest from the last meeting: 593):
596 selftests: `mptcp_join`: fastclose also generate "plain" RST
[bug] [selftests] @pabeni:
- See patches above
595 selftests: `mptcp_connect`: MPTCP -> MPTCP timeout during closing
time [bug] [selftests]:
- tried to reproduce the issue but no luck so far.
- It is possible multiple data fin can be sent with different
sequences: in this case, we might ignore the following data_fin
- Matth: TODO: check previous packet traces when that might have
happened → done
594 selftests: `mptcp_connect`: MPTCP -> TCP timeout during closing
time [bug] [selftests]:
- tried to reproduce with a packetdrill test
- maybe try with no delay: it will depend on when the worker
processes the FIN
- and also read the last bytes carried in the FIN
- and also injecting more data without reading them
- TODO Matth: check ↑ → done
Recently discussed
592 Lower throughput with MPTCP vs TCP [bug]:
- HW GRO seems to be the culprit: it would be good to start
discussing with NICs vendors (NVidia, Intel, etc.). It **might** be easy
to add some exceptions for MPTCP in some NIC firmwares. Who has contact
and/or could help to find the right people to talk to?
- Strange TCP is not even higher, maybe because the limitation for
TCP was the IO? (check with 'iperf3 -Z' and/or iotop) If yes, maybe we
can never do better than TCP connections in this case. Or then the test
case is maybe not realistic: it might be different with many
applications and all sort of data to different hosts? Or maybe only the
resilience is important here, and doing as good as TCP is fine?
- It would be interesting to do the same validations on top of
Paolo's pending series (EDIT: one left as of today, the rest is in our
`export` branch), because there are new optimisations, especially when
the latency is very short like what you would have on the same host or
in a DC.
591 in-kernel PM: not established paths when used with IPv6
link-local addresses [bug]:
- Can be closed?
- TODO Matth → Done
588 in-kernel PM: support NUMA awareness [enhancement] [feedback-needed]:
- Maybe BPF PM is the solution
549 pm: in-kernel: set `msk->pm.work_pending` back to `true` [bug]
[selftests] [pm]:
- New comment from Gang: to be checked
Bugs (opened, flagged as "bug" and assigned)
596 selftests: `mptcp_join`: fastclose also generate "plain" RST
[bug] [selftests] @pabeni
575 sockopt: support `TCP_MD5SIG` and `TCP_MD5SIG_EXT` for
transparent integrations [bug] @geliangtang
498 in-kernel PM: closed subflows before `RM_ADDR` will not decrement
`add_addr_accepted` [bug] [pm] @geliangtang
484 BPF: `setsockopt` on an MPTCP socket: check support [bug] [bpf]
@geliangtang
445 MPTCP Info counters: start seq numbers ones from 0 [bug] @geliangtang
403 userspace pm: only remove entry from `local_addr_list` when
sending a `REMOVE_ADDR` [bug] [pm] @geliangtang
343 scheduler: react when subflow-level events pop up (ACK/RTO) [bug]
[sched] @geliangtang
Bugs (opened and flagged as "bug" and not assigned)
595 selftests: `mptcp_connect`: MPTCP -> MPTCP timeout during closing
time [bug] [selftests]
594 selftests: `mptcp_connect`: MPTCP -> TCP timeout during closing
time [bug] [selftests]
592 Lower throughput with MPTCP vs TCP [bug]
591 in-kernel PM: not established paths when used with IPv6
link-local addresses [bug]
587 in-kernel PM: Server is only using single IP to establish
subflows [bug]
586 [syzkaller] WARNING in `subflow_data_ready` [bug] [syzkaller]
583 KMemLeak: from `mptcp_sk_clone_init()` [bug] [selftests]
581 Default scheduler under-utilizes newly restored subflows on
long-lived flows after WAN failback [bug]
574 [selftests] `mptcp_connect.sh`: `copyfd_io_poll`: poll timed out
[bug] [selftests]
572 [syzkaller] WARNING in `skb_try_coalesce` [bug] [syzkaller]
567 [syzbot] WARNING in `__mark_subflow_endp_available` [bug]
[syzkaller] [reproducer]
561 [syzkaller] WARNING in `mptcp_pm_alloc_anno_list` [bug] [syzkaller]
549 pm: in-kernel: set `msk->pm.work_pending` back to `true` [bug]
[selftests] [pm]
508 Divide error on device removal [bug] [feedback-needed]
505 Subflows won't be used when the interface for the main flow is a
subinterface [bug] [feedback-needed]
495 [syzkaller] KASAN: slab-use-after-free Read in
`__inet6_lookup_established` [bug] [syzkaller]
494 [syzkaller] WARNING in `__skb_flow_dissect` [bug] [syzkaller]
492 userspace pm: subflows counter: not updated when a subflows is
removed from the other end [bug]
491 selftests: `mptcp_join`: tests `Infinite map` & `MP_FAIL MP_RST`
are unstable [bug] [selftests]
483 [6.1] selftests: `mptcp_connect.sh` failures with
`copyfd_io_poll: poll timed out` errors [bug] [selftests]
476 packetdrill: coalescing feature doesn't take into account MPTCP
options [bug] [packetdrill]
475 `simult_flows`: unbalanced bwidth tests are unstable [bug]
[selftests]
439 mptcp vs net.ipv4.tcp_shrink_window [bug]
336 Issues with BPF packet scheduler [bug] [sched] [bpf]
325 selftests: `mptcp_join`: test 18 `signal addresses race test` is
sometimes failing [bug] [selftests]
324 selftests: `mptcp_join`: tests `fastclose test` & `fastclose
server test` are sometimes failing [bug] [selftests]
315 Issues with backup flow [bug]
In Progress (opened, new feature and assigned)
568 socket option: support `IPV6_TCLASS` [enhancement] @geliangtang
480 MPTCP and KTLS support [enhancement] @geliangtang
434 in-kernel PM: increase subflows limit [enhancement] [pm] @matttbe
353 sockopt: uniform code to get/set values [enhancement] @matttbe
349 scheduler: (un)mark a subflow as "stale" [enhancement] [sched]
@geliangtang
296 Support IP_TTL & IPV6_UNICAST_HOPS [enhancement] @matttbe
167 packetdrill: add coverage for RM_ADDR [enhancement] [packetdrill]
@dcaratti
75 BPF: packet scheduler [enhancement] [sched] [bpf] @geliangtang
74 BPF: path manager [enhancement] [pm] [bpf] @geliangtang
Assigned Questions (opened, questions and assigned)
548 disable mptcp for an interface [question] @matttbe
Open questions (opened, questions and not assigned)
562 [performance] How much is the overhead of MPTCP？ [question]
530 about mptcp subflow creation question [question] [feedback-needed]
520 Question about proper MPTCP configuration over VPN. [question]
[feedback-needed]
412 shadowsocks-libev not working, need guidance for PM configuration
[question] [feedback-needed]
For later (opened and not assigned)
590 in-kernel PM: use either `subflow` or `signal` depending on
connector/listener app [enhancement]
588 in-kernel PM: support NUMA awareness [enhancement] [feedback-needed]
584 [syzkaller] improve coverage [syzkaller]
578 sockopt: support `SO_MAX_PACING_RATE` to limit the whole MPTCP
connection [enhancement]
566 pm: use `signal` + `subflow` + `fullmesh` [enhancement] [pm]
556 Implement all required bits to C/R MPTCP connections from
user-space (CRIU) [enhancement]
552 pm: in-kernel: delay before creating new subflows [enhancement]
545 Deal with multiple pure TCP ACK without MPTCP options [enhancement]
543 TCPDump: display `MP_TCPRST` info [enhancement]
542 in-kernel PM: limit to one subflow per network interface for each
connection [enhancement] [pm]
535 TCPDump: display `MP_CAPABLE` C flag [enhancement]
525 tests: cover more MPTCP socket options [enhancement]
[packetdrill] [selftests]
521 Implement psock_update_sk_prot, for eBPF SOCKMAP support
[enhancement]
519 Reset the connection once "MPTCP has been verified" instead of a
fallback [enhancement]
511 Add MIB counters for sent `MPTCP_RST_EMPTCP` [enhancement]
497 in-kernel PM: check to create new subflows when some are
destroyed [enhancement] [pm]
496 in-kernel PM: store received `ADD_ADDR` [enhancement] [pm]
481 remove refcount on TCP sockets for the timers [enhancement]
467 Scheduler: add redundant scheduler support in BPF [enhancement]
[sched]
462 selftests: print (and set?) random seed [enhancement] [selftests]
461 Protocol: extend Data-Level Length field [enhancement] [protocol]
448 Allow sending ADD_ADDR IPv6 with port-number if TCP-timestamps
are enabled [enhancement]
441 optimize established options generation [enhancement]
440 in-kernel PM: re-establish subflows after "network" errors
[enhancement] [pm]
426 selftests: validate listener sockets with multiple accepts
[enhancement] [selftests]
396 ci: periodically run selftests from the export branch on stable
kernels [enhancement] [selftests]
389 packetdrill: increase coverage for "early `close()`" cases
[enhancement] [packetdrill]
350 scheduler: API changes (tasks) [enhancement] [sched]
348 scheduler: frequently probe "stale" subflow with reinjected data
[enhancement] [sched]
345 scheduler: "penalise" some subflows by sending less than their
cwnd [enhancement] [sched]
344 scheduler: new callback to initiate optimisations [enhancement]
[sched]
337 in-kernel PM: listen socket: support "behind a NAT" use case
[enhancement] [pm]
335 Server to Client subflows: option to allow joining the initial
address and port [enhancement] [pm]
334 in-kernel PM: send >1 ADD_ADDR before the establishment of a new
subflow [enhancement] [pm]
332 scheduler: implement a "opportunistic retransmission"
[enhancement] [sched]
298 CG Memory accounting for passive MPTCP socket [enhancement]
289 mptcp: audit comments for non inclusive language [enhancement]
273 Packetdrill: test outgoing MP_FASTCLOSE [enhancement] [packetdrill]
266 Packetdrill: add MP_FAIL coverage [packetdrill]
236 Review supported sockopts list [enhancement]
234 Packetdrill: Support MPC+DATA+checksum error [enhancement]
[packetdrill]
222 Netlink event API: add SUBFLOW_CREATED event [enhancement]
215 TCP Urgent pointer and MPTCP [enhancement]
213 add MPTCP man page [enhancement]
208 better handing of ssk memory pressure in the TX path [enhancement]
202 Add sendmsg support for ancillary data [enhancement]
197 more mibs needed [enhancement]
180 Get an update when MPTCP fall back to TCP [enhancement]
177 improve retransmit subflow selection [enhancement]
169 packetdrill: add coverage for ADD_ADDR and MP_JOIN on a different
port [enhancement] [packetdrill]
150 remove completely workqueue usage [enhancement]
141 avoid acquiring mptcp_data_lock() twice in the receive path
[enhancement]
78 notify the application (userspace) when a subflow is
added/removed [enhancement]
Recently closed (since 2025-10-22)
593 New MPTCP connections won't be created if there's no connectivity
on primary subflow [question]:
- Could be possible to handle that on MPTCP side, but it doesn't
seem easy, and it might break expectations from userspace. Better to
handle that with routes.
589 selftests: `simult_flows`: "unbalanced bwidth with opposed,
unbalanced delay" is unstable [bug] [selftests] @geliangtang
Packetdrill PRs
181 22 hours ago matttbe mptcp: new DATA_FIN retrans tests:
- New test, passing, can be merged if reviews are OK
180 12 days ago matttbe mptcp: unskip TCP_NOTSENT_LOWAT test:
- Seems passing now, can be merged if reviews are OK
162 4 months ago geliangtang mptcp: add splice test:
- TODO: Matth: check with Geliang if he plans to update it → Done
52 9 months ago dcaratti Add initial coverage for RM_ADDR
mptcp.dev PRs
/
mptcpd PRs
327 21 days ago matttbe man: plugin-dir: not world writable
326 1 hour ago matttbe flags: add 'laminar' endpoints support
323 27 days ago matttbe interface: add `deny_join_id0` parameter
FYI: Current Roadmap:
- Bugs:
https://github.com/multipath-tcp/mptcp_net-next/issues?q=is%3Aissue+is%3Aopen+label%3Abug
- Current/Coming merge window (6.19):
https://github.com/orgs/multipath-tcp/projects/17
- For later: https://github.com/orgs/multipath-tcp/projects/1
Patches to send to NetDev:
- Fixes for other trees:
- [f35023f92ed5] selftests: bpf: fix recent compilation errors
(Matthieu Baerts (NGI0))
- [a7cb48db2368] gcov: add support for GCC 15 (Peter Oberparleiter)
- Fixes for -net:
- [fa4c16e5903b] selftests: mptcp: connect: fix fallback note
due to OoO (Matthieu Baerts (NGI0))
- [b00cc0dfd3ef] selftests: mptcp: join: rm: set backup flag
(Matthieu Baerts (NGI0))
- [db4e86a2e6c2] selftests: mptcp: join: endpoints: longer
transfer (Matthieu Baerts (NGI0))
- [732cf2c50ddb] selftests: mptcp: join: userspace: longer
transfer (Matthieu Baerts (NGI0))
- [eb98322c2d06] selftests: mptcp: join: fastclose: drop plain
RST (Matthieu Baerts (NGI0)):
- Will be dropped soon
- [604f4ade5c21] selftests: mptcp: connect: trunc: read all recv
data (Matthieu Baerts (NGI0))
- [3c6e894585f2] mptcp: fix address removal logic in
mptcp_pm_nl_rm_addr (Gang Yan)
- Fixes for net-next: /
- Features for net-next:
- [8aa057db1bba] mptcp: pm: init and release mptcp_pm_ops
(Geliang Tang)
- [fd8a6842c53a] mptcp: pm: add get_local_id() interface
(Geliang Tang)
- [07d670ca1d24] mptcp: pm: add get_priority() interface
(Geliang Tang):
- blocked by pm->lock refactoring that is needed
- [ffaa4e7399dc] mptcp: cleanup fallback data fin reception
(Paolo Abeni)
- [8c97e58c6aea] mptcp: cleanup fallback dummy mapping
generation (Paolo Abeni):
- there might be a fix for that
- [6628958a45a2] mptcp: ensure the kernel PM does not take
action too late (Paolo Abeni)
- [692254038cb1] mptcp: do not miss early first subflow close
event notification (Paolo Abeni)
- [9f8782509d81] mptcp: make mptcp_destroy_common() static
(Paolo Abeni)
- [440bbc559b86] mptcp: drop the __mptcp_data_ready() helper
(Paolo Abeni)
- [6148a584cce9] mptcp: handle first subflow closing
consistently (Paolo Abeni)
- [6d5d949d5258] mptcp: borrow forward memory from subflow
(Paolo Abeni)
- [7984871487f7] mptcp: introduce mptcp-level backlog (Paolo Abeni)
- [8afd86c5dc49] mptcp: leverage the backlog for RX packet
processing (Paolo Abeni):
- wait for the fix
- [5d9a58722e9c] selftests: mptcp: lib: introduce
'nstat_{init,get}' (Matthieu Baerts (NGI0))
- [e5aad6ad52b4] selftests: mptcp: lib: remove stats files args
(Matthieu Baerts (NGI0))
- [0cdbeae681cb] selftests: mptcp: lib: stats: remove nstat rate
columns (Matthieu Baerts (NGI0))
- [e500e971e1bb] selftests: mptcp: join: dump stats from history
(Matthieu Baerts (NGI0))
- [22e437406bc4] selftests: mptcp: lib: get counters from nstat
history (Matthieu Baerts (NGI0)):
- should be sent after the ones for net + merge with net-next
- Features for net-next (next): /
- Features for other trees:
- [be06d52e341d] bpf: Add mptcp_subflow bpf_iter (Geliang Tang)
- [599fef9658dd] selftests/bpf: More endpoints for endpoint_init
(Geliang Tang)
- [4f16da3ca774] selftests/bpf: Drop cgroup_fd of run_mptcpify
(Geliang Tang)
- [2eff5ee21d70] bpf: Add mptcp packet scheduler struct_ops
(Geliang Tang)
- [0d54dd03ab0d] bpf: Export mptcp packet scheduler helpers
(Geliang Tang)
- [2aca918a8b6e] selftests/bpf: Add bpf scheduler test (Geliang
Tang)
- [c4d2d29ba444] selftests/bpf: Add bpf_first scheduler & test
(Geliang Tang)
- [b65f96d93233] selftests/bpf: Add bpf_bkup scheduler & test
(Geliang Tang)
- [84e47205025f] selftests/bpf: Add bpf_rr scheduler & test
(Geliang Tang)
- [df0031f69a5c] selftests/bpf: Add bpf_red scheduler & test
(Geliang Tang)
- [b81ba7f0c095] selftests/bpf: Add bpf_burst scheduler & test
(Geliang Tang):
- waiting for a refactoring of the sched API
Extra tests:
- news about Syzkaller? (Matth):
- Some issues there, but not reproduceable:
https://syzkaller.appspot.com/upstream/s/mptcp
- maybe: a memory corruption elsewhere?
- not reproduced for one month
- Packetdrill (Davide):
- /
- Patchew (Davide):
- Was down this WE, but fixed by Davide
- CI (Matth):
- https://ci-results.mptcp.dev/flakes.html
- https://netdev.bots.linux.dev/flakes.html?tn-needle=mptcp
- /
Next meeting:
- Next one on Wednesday, the 19th of November.
- 17:30 UTC (8.30am PDT, 5.30pm CEST, 11.30pm CST)
- Still open to everyone!
- https://annuel2.framapad.org/p/mptcp_upstreaming_20251119
Feel free to comment on these topics and suggest new ones for the next
meeting!
Talk to you soon,
Matt
--
Sponsored by the NGI0 Core fund.
next reply other threads:[~2025-11-12 17:58 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-12 17:58 Matthieu Baerts [this message]
2025-11-13 7:57 ` [Weekly meetings] MoM - 5th of November 2025 Geliang Tang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=5e7430ff-ef27-49ef-a295-875a8c6fdcd1@kernel.org \
--to=matttbe@kernel.org \
--cc=mptcp@lists.linux.dev \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox