* [Weekly meetings] MoM - 15th of April 2021
@ 2021-04-16 7:43 Matthieu Baerts
0 siblings, 0 replies; only message in thread
From: Matthieu Baerts @ 2021-04-16 7:43 UTC (permalink / raw)
To: MPTCP Upstream
Hello everyone,
Yesterday, we had our 144th meeting with Mat and Ossama (Intel), Florian
and Paolo (RedHat), Geliang (Xiaomi), Maxim Galaganov (Parents helper)
and myself (Tessares).
Thanks again for this new good meeting!
Here are the minutes of the meeting:
netdev (if mptcp ML is in cc) (/):
/
our repo (by: Florian Westphal, Geliang Tang, Nico Pache):
1466512 [mptcp-next] mptcp: fix format specifiers for unsigned int
1465840 [mptcp-next,v2,10/10] selftests: mptcp: add packet mark test case
1465839 [mptcp-next,v2,09/10] mptcp: sockopt: add TCP_CONGESTION and
TCP_INFO
1465838 [mptcp-next,v2,08/10] mptcp: setsockopt: SO_DEBUG and no-op
options
1465837 [mptcp-next,v2,07/10] mptcp: setsockopt: add SO_INCOMING_CPU
1465836 [mptcp-next,v2,06/10] mptcp: setsockopt: add SO_MARK support
1465835 [mptcp-next,v2,05/10] mptcp: setsockopt: support SO_LINGER
1465834 [mptcp-next,v2,04/10] mptcp: setsockopt: handle receive/send
buffer a...
1465833 [mptcp-next,v2,03/10] mptcp: setsockopt: handle SO_KEEPALIVE
and SO_P...
1465832 [mptcp-next,v2,02/10] mptcp: tag sequence_seq with socket state
1465831 [mptcp-next,v2,01/10] mptcp: add skeleton to sync msk socket
options ...
1462712 [mptcp-next,2/2] mptcp: use mptcp_for_each_subflow in mptcp_close
1466086 [v2,5/6] kunit: mptcp: adhear to KUNIT formatting standard:
- /!\ a KConfig option has been renamed:
- MPTCP_KUNIT_TESTS → MPTCP_KUNIT_TEST
Pending patches:
- The list of pending patches can be seen on PatchWork:
https://patchwork.ozlabs.org/project/mptcp/list/?state=*
netdev (if mptcp ML is in cc) (by: /):
/
our repo (by: Florian Westphal, Geliang Tang, Mat Martineau, Paolo
Abeni, Yonglong Li):
1370700: RFC: [RFC,2/4] tcp: move selected mptcp helpers to tcp.h/mptcp.h
1370702: RFC: [RFC,4/4] tcp: parse tcp options contained in reset packets:
- WIP
1375893: RFC: [RFC,mptpcp-next] mptcp: add ooo prune support:
- WIP
1395128: RFC: [1/5] tcp: make two mptcp helpers available to tcp stack
1395133: RFC: [5/5] mptcp: send fastclose if userspace closes socket
with unread data:
- WIP
1426554: Changes Requested: [PATCHi,iproute2] mptcp: add support for
event monitoring:
- WIP
- When is their window closing?
1450496: RFC: [RFC,2/2] mptcp: add MP_FAIL support:
- WIP
1459518: Changes Requested: [v2,mptcp-next,01/16] mptcp: add
csum_enabled in mptcp_sock
1459520: Changes Requested: [v2,mptcp-next,02/16] mptcp: generate the
data checksum
1459519: Changes Requested: [v2,mptcp-next,03/16] mptcp: add csum_reqd
in mptcp_out_options
1459521: Changes Requested: [v2,mptcp-next,04/16] mptcp: send out
checksum for MP_CAPABLE with data
1459522: Changes Requested: [v2,mptcp-next,05/16] mptcp: send out
checksum for DSS
1459523: Changes Requested: [v2,mptcp-next,06/16] mptcp: add csum_reqd
in mptcp_options_received
1459524: Changes Requested: [v2,mptcp-next,07/16] mptcp: add sk
parameter for mptcp_parse_option
1459525: Changes Requested: [v2,mptcp-next,08/16] mptcp: receive
checksum for MP_CAPABLE with data
1459526: Changes Requested: [v2,mptcp-next,09/16] mptcp: receive
checksum for DSS
1459527: Changes Requested: [v2,mptcp-next,10/16] mptcp: validate the
data checksum
1459528: Changes Requested: [v2,mptcp-next,11/16] mptcp: add the mib for
data checksum
1459529: Changes Requested: [v2,mptcp-next,12/16] mptcp: add a new
sysctl checksum_enabled
1459530: Changes Requested: [v2,mptcp-next,13/16] mptcp: add
mptcpi_csum_enabled in mptcp_info
1459531: Changes Requested: [v2,mptcp-next,14/16] mptcp: add trace event
for data checksum
1459532: Changes Requested: [v2,mptcp-next,15/16] selftests: mptcp:
enable checksum in mptcp_connect.sh
1459533: Changes Requested: [v2,mptcp-next,16/16] selftests: mptcp:
enable checksum in mptcp_join.sh:
- seems more complex than expected
- critical part is queuing the data from the subflow received queue
with DSS granularity → we need a full DSS coverage before validating the
checksum
- Paolo is helping Geliang
1460888: Changes Requested: [v3] mptcp: ensure there is unacked data at
all subflow:
- it seems we are not going to have a v4 there
- we can archive it
- follow-up ticket:
https://github.com/multipath-tcp/mptcp_net-next/issues/177
1462554: RFC: [RFC,mptcp-next] mptcp: Retransmit DATA_FIN:
- Mat is working on a v1
1466292: Changes Requested: [mptcp-next] mptcp: ignore unsupported msg
flags:
- Some discussions on the ML with Florian
- v2 expected
- v2 will conflict with MSG_PEEK patch ↓ but not big conflicts and
easy to resolve
1466461: Changes Requested: [v2,1/2] mptcp: add MSG_PEEK support
1466462: Changes Requested: [v2,2/2] selftests: mptcp: add a test case
for MSG_PEEK:
- Review has been done by Paolo
Issues on Github:
https://github.com/multipath-tcp/mptcp_net-next/issues/
Recently opened (latest from last week: 178)
/
Bugs (opened, flagged as "bug" and assigned)
178 Server hang when proxying over multiple subflows [bug] [v5.11]
[v5.12]:
- Maxim reported a nice and complete bug report
- Paolo suggested a workaround that seems to avoid the issue
- But the patch is not "upstreamable" → will increase mem usage
- We need to understand the root cause
- Probably a use-after-free but hidden to KASAN (pfrag)
- Paolo is looking at a patch adding some poisoning hoping to
have KASAN finding the issue
- After meeting note: What could also help:
- strace of the server before the crash
176 BPF selftest got "fallback to TCP" and "Attempt to release TCP
socket in state 8" errors [bug] @geliangtang
146 DATA_FIN is not retransmitted on timeout [bug] @mjmartineau:
- See above
Bugs (opened and flagged as "bug" and not assigned)
172 WARNING in sk_stream_kill_queues [bug]
162 sendmsg()/recvmsg() fail when an unknown CMSG argument is
provided [bug]
137 selftests: simult_flows.sh: unbalanced bwidth tests are unstable
[bug]
120 [interop] netnext is dropping packets, causing MPTCP-level
retransmissions on mptcp.org [bug]
107 Review use of WARN_ON() / WARN_ON_ONCE() [bug]
65 clearing properly the status in listen() [bug]
56 msk connection state set without msk lock [bug]
In Progress (opened and assigned):
167 packetdrill: add coverage for RM_ADDR [enhancement]
[packetdrill] @dcaratti
158 iproute2: change backup mode (MP_PRIO) for active connections
[enhancement] [iproute2] @dcaratti
143 Packetdrill: ADD_ADDR for v6 only socket should only contain v6
addresses [enhancement] [packetdrill] @dcaratti
134 Checksum support [enhancement] @geliangtang
96 Python: add support for IPPROTO_MPTCP [enhancement] @matttbe
53 MP_TCPRST support [enhancement] @fw-strlen
52 MP_FAIL support [enhancement] @geliangtang
Recently closed (since last week)
None.
FYI: Current Roadmap:
- Bugs: https://github.com/multipath-tcp/mptcp_net-next/projects/2
- Current/Coming merge window (5.13):
https://github.com/multipath-tcp/mptcp_net-next/projects/8
- For later: https://github.com/multipath-tcp/mptcp_net-next/projects/4
Patches to send to netdev:
- net:
- /
- net-next:
- Maybe the last week to push stuff
- 3 series are pending:
- sockopt from Paolo:
- They have been stressed by syzkaller
- *@Christoph:* nothing to report?
- tracepoint support from Geliang:
- seems OK to be sent upstream
- sockopt from Florian:
- seems OK to be sent? To be stressed first maybe?
- Should be OK without
- can be sent with Paolo's series
Extra tests:
- news about Syzkaller? (Christoph):
- /
- news about interop with mptcp.org? (Christoph):
- /
- news about Intel's kbuild? (Mat):
- some progress but no summary email yet
- apparently no issues on our side
- packetdrill (Davide):
- /
- CI (Matth):
- The notification on IRC is currently reporting two things:
- Task successfully finished! → it means no crash,
compilation error, global timeout
- tests_ok → current broken: it will always report false
- But we can check on the provided link to open the
"summary" file and see the result
- *Matth* will look at that
mptcpd: new release: v0.7:
- https://github.com/intel/mptcpd/releases/tag/v0.7
- MPTCP path management generic netlink events recently added to
the upstream Linux kernel are now supported, and propagated to mptcpd
plugins. The same API is also found in the multipath-tcp.org kernel.
Differences between the two kernels are transparent to mptcpd plugins.
- Mptcpd now supports the upstream Linux kernel server-oriented
MPTCP_PM_CMD_SET_FLAGS generic netlink command through the new
mptcpd_pm_set_flags() function. It allows the user to set MPTCP flags
for a specific local IP address.
- An incorrect buffer size calculation when calling
mptcpd_pm_add_addr() when using an IPv6 addresses on platforms running
the upstream Linux kernel was corrected.
- An "operation not supported" error that occurred when attempting
to retrieve MPTCP resource limits through the mptcpd_pm_get_limits()
function was fixed.
- A network interface lookup problem in the "sspi" plugin was
corrected.
- Building mptcpd against ELL 0.39 is now supported.
Load Balancer:
- Florian would like to improve LB's support
- Any paper/thesis/... to look at to know what's important to support?
- Or feedback from existing deployments?
- *@Christoph* maybe? :)
- https://www.rfc-editor.org/rfc/rfc8684.html#section-3.1-20.6
-
https://datatracker.ietf.org/doc/draft-duchene-mptcp-load-balancing/01/
- If each server behind the firewall can have a dedicated IP that
can be announced, we can have a stateless LB on the "initial subflow"
side. No LB needed on the other IP, only accepting MPJoin
- Still could be good to have ConnTrack being able to say that one
flow (e.g. MPJ) is belonging to another connection.
Next meeting:
- On Thursday, the 22th of April.
- Usual UTC time: 15:00 UTC (8am PDT, 5pm CEST, 11pm CST)
- Still open to everyone!
- https://annuel2.framapad.org/p/mptcp_upstreaming_20210422
Feel free to comment on these points and propose new ones for the next
meeting!
Talk to you on Thursday,
Matt
--
Tessares | Belgium | Hybrid Access Solutions
www.tessares.net
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2021-04-16 7:43 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-04-16 7:43 [Weekly meetings] MoM - 15th of April 2021 Matthieu Baerts
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).