[MPTCP] [Weekly meetings] MoM - 3rd of September 2020

[MPTCP] [Weekly meetings] MoM - 3rd of September 2020

[Matthieu Baerts]

[-- Attachment #1: Type: text/plain, Size: 9215 bytes --]

Hello everyone,

Last Thursday, we had our 114th meeting with Mat (Intel OTC), Christoph 
(Apple), Paolo and Davide (RedHat), Geliang (Xiaomi) and myself (Tessares).

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.ozlabs.org/project/mptcp/list/?state=3

     netdev (if mptcp ML is in cc) (Davide Caratti, YueHaibing):

1355689  [net-next] selftests: mptcp: fix typo in mptcp_connect usage
1353662  [net-next] mptcp: Remove unused macro MPTCP_SAME_STATE


     our repo (by: Florian Westphal, Paolo Abeni):

1353755  fix for most poll selftest timeouts
1352466  [v2] Squash-to: "mptcp: cleanup mptcp_subflow_discard_data()"
1349374  [mptcp-next,1/1] mptcp: fix stale subflow->writeable caching
1349247  [mptcp-next] mptcp: free acked data before waiting for more memory
1349244  [mptcp-next] mptcp: only wake parent if subflow is writable
1348371  [mptcp-next,v3] mptcp: adjust mptcp receive buffer limit if 
subflow h...



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: Nicolas Rybowski):

1349344  [bpf-next,3/3] bpf: add 'bpf_mptcp_sock' structure and helper
1349341  [bpf-next,2/3] mptcp: attach subflow socket to parent cgroup


     our repo (by: Florian Westphal, Geliang Tang, Matthieu Baerts, 
Nicolas Rybowski, Paolo Abeni):

1325171: Changes Requested: [mptcp-next] selftests: mptcp: interpret \n 
as a new line:
     - TODO: Matth

1349790: Changes Requested: [mptcp-next] mptcp: keep track of receivers 
advertised window:
     - Florian is working on a v2

1353032: Awaiting Upstream: [v7,mptcp-next,1/4] mptcp: send out ADD_ADDR 
with echo flag
1353033: Awaiting Upstream: [v7,mptcp-next,2/4] mptcp: add ADD_ADDR 
related mibs
1353034: Awaiting Upstream: [v7,mptcp-next,3/4] selftests: mptcp: add 
ADD_ADDR mibs check function
1353035: Awaiting Upstream: [v7,mptcp-next,4/4] mptcp: add 
accept_subflow re-check:
     - TODO: Matth → done

1355637: New: [v9,mptcp-next,1/8] mptcp: remove addr and subflow in PM 
netlink
1355638: New: [v9,mptcp-next,2/8] mptcp: fix 
mptcp_pm_nl_rm_addr_received logic issue
1355639: New: [v9,mptcp-next,3/8] mptcp: implement mptcp_pm_remove_subflow
1355643: New: [v9,mptcp-next,4/8] mptcp: fix every subflow's local_id is 
zero
1355644: New: [v9,mptcp-next,5/8] mptcp: fix subflow's remote_id is zero 
issue
1355645: New: [v9,mptcp-next,6/8] mptcp: add RM_ADDR related mibs
1355646: New: [v9,mptcp-next,7/8] selftests: mptcp: add remove cfg in 
mptcp_connect
1355647: New: [v9,mptcp-next,8/8] selftests: mptcp: add remove addr and 
subflow test cases:
     - Paolo will look at that
     - v10 will arrive next week

1355829: Awaiting Upstream: [mptcp-next] mptcp: use _fast lock version 
in __mptcp_move_skbs:
     - TODO: Matth → Done
     - Florian was right, as usual :-)

1355957: New: [v2] tcp: propagate MPTCP skb extensions on xmit splits:
     - TODO: Matth → Done

1355993: Awaiting Upstream: [mptcp-next,v2,1/5] bpf: expose is_mptcp 
flag to bpf_tcp_sock
1355994: Awaiting Upstream: [mptcp-next,v2,2/5] mptcp: attach subflow 
socket to parent cgroup
1355996: Awaiting Upstream: [mptcp-next,v2,3/5] bpf: add 
'bpf_mptcp_sock' structure and helper
1355997: Awaiting Upstream: [mptcp-next,v2,4/5] bpf:selftests: add MPTCP 
test base
1355998: Awaiting Upstream: [mptcp-next,v2,5/5] bpf:selftests: add 
bpf_mptcp_sock() verifier tests:
     - TODO: Matth → done
     - Then can be sent upstream

1356477: New: Squash-to: "mptcp: add the incoming RM_ADDR support":
     - needs review



Issues on Github:
     https://github.com/multipath-tcp/mptcp_net-next/issues/

     Recently opened (latest from last week: 82)

   84  Received DATA_FIN+FIN does not trigger wakeup [bug] v5.9 
@mjmartineau ‡:
       - when checking 5.8 (Fedora) vs 5.9-rc3 (-net master branch)
       - because of a FIN without data payload
       - Mat is looking at that
       - Could be related to an issue seen by Paolo

   83  [syzkaller] No output from test machine:
       - Christoph will try to reproduce it
       - it means: the VM died and disappeared
       - there is a reproducer with an MPTCP socket
       - not sure if it is linked to MPTCP code


     Bugs (opened, flagged as "bug" and assigned)

   84  Received DATA_FIN+FIN does not trigger wakeup [bug] v5.9 
@mjmartineau ‡:
       - see above

   81  packetdrill detects mismatching 'B' flag in mp_join_client.pkt 
[bug] @dcaratti:
       - Should be addressed with PR#18 in Packetdrill
       - Matth: TODO → review this PR#18 and merge (or Davide can 
self-merge it)

   69  Packetdrill: dss: failing with new DATA_FIN patches [bug] @dcaratti:
       - Should be addressed with PR#17 in Packetdrill
       - Matth: TODO → review this PR#17 and merge (or Davide can 
self-merge it)


     Bugs (opened and flagged as "bug" and not assigned)

   73  Simultaneous xmit issue [bug]
       - can be closed
       - Matth: TODO update ticket

   71  [interop] Bad mapping: ssn=16194949 map_seq=16120693 
map_data_len=31416 [bug] [interop]
       - can be closed after an ACK from @Christoph
       - Matth: TODO update ticket

   70  [syzkaller] WARNING in mptcp_reset_timer [bug] [syzkaller]
   65  clearing properly the status in listen() [bug]

   62  [syzkaller] WARNING in __mptcp_move_skbs_from_subflow [bug] 
[syzkaller]:
       - Matth: TODO update ticket

   56  msk connection state set without msk lock [bug]


     In Progress (opened and assigned)

   82  improve packet scheduler for asymmetric links [enhancement] @pabeni :
       - after having tweaked the scheduler, links with different 
conditions can be fully used
       - having an issue but more experiments are needed

   55  ADD_ADDR: IPv6 support [enhancement] @geliangtang
   50  REMOVE_ADDR support [enhancement] @geliangtang
   49  ADD_ADDR: echo bit support [enhancement] @geliangtang :
       - see patches above

   43  [syzkaller] Change syzkaller to exercise MPTCP inet_diag 
interface [enhancement] [syzkaller] @cpaasch :
       - WIP


     Recently closed (since last week)

   67  `./mptcp_connect.sh -m mmap` test blocks [bug]
   17  allow non 'backup' subflows creation [enhancement] @pabeni



Patches from the "export" branch:
     - multiple xmit:
         - we would need to wait for the selftests to be enabled → 
linked to #82
         - alternative: enable selftests for symmetric links and fixes 
later?
         - will see how it is going in the next weeks with #82

     - REMOVE_ADDR: could be sent when the last patch (squashed-to) is 
integrated

     - selftests/mptcp: Better delay & reordering configuration:
         - Christoph can send that



New features for net-next:
     - continue to let the different authors to send patches from the 
"export" branch to net-next?
     - Or maintain an mptcp-next branch sync with net-next and ask for a 
pull?
     - at the end, that's more for the net maintainers. Maybe at some 
points they will suggest us to send PR instead of individual patches?
     - Should we start discussions with them about that? Or just wait?
     - no decision yet, we continue like before



FYI: Current Roadmap:
     - Bugs: https://github.com/multipath-tcp/mptcp_net-next/projects/2
     - Current merge window (5.10): 
https://github.com/multipath-tcp/mptcp_net-next/projects/5
     - For later: https://github.com/multipath-tcp/mptcp_net-next/projects/4



Extra tests:
     - news about Syzkaller? (Christoph):
         - see Github section

     - news about interop with mptcp.org? (Christoph):
         - /

     - news about Intel's kbuild? (Mat):
         - we got a report on the ML (unpacking failed) but doesn't seem 
related to MPTCP
         - selftests are still unstable, seems related to the setup, not 
MPTCP code
         - not sure any more what is running: also testing the new 
selftests?
         - Mat will monitor that

     - packetdrill (Davide):
         - planning to work on:
             - DATA_FIN: to cover this feature better (see GH discussions)
             - covering MP_PRIO

     - CI (Matth):
         - packetdrill tests are failing
         - maybe linked to #84? → No, see #85
         - Matth: TODO → take traces on "any" → DONE, see #85
         - Matth: continue the "git bisect"
         - Davide has issues only with MP_JOIN (src addr)



Next meeting:
     - We propose to have the next meeting on Thursday, the 10th of 
September.
     - Usual time: 15:00 UTC (8am PDT, 5pm CEST, 11pm UTC+8)
     - Still open to everyone!
     - https://annuel2.framapad.org/p/mptcp_upstreaming_20200910



Feel free to comment on these points and propose new ones for the next
meeting!

Talk to you next week,
Matt
-- 
Tessares | Belgium | Hybrid Access Solutions
www.tessares.net