[MPTCP] [Weekly meetings] MoM - 8th of October 2020

[MPTCP] [Weekly meetings] MoM - 8th of October 2020

[Matthieu Baerts]

[-- Attachment #1: Type: text/plain, Size: 8827 bytes --]

Hello everyone,

Last Thursday, we had our 119th meeting with Mat and Ossama (Intel OTC), 
Paolo and Davide (RedHat) 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) (Matthieu Baerts, Paolo Abeni, Rikard 
Falkeborn):

1377527  [net-next] selftests: mptcp: interpret \n as a new line
1377217  [net-next] mptcp: fix infinite loop on recvmsg()/worker() race.
1376734  [net-next] mptcp: don't skip needed ack
1376726  [net] mptcp: more DATA FIN fixes
1376565  [net-next,v2,1/2] mptcp: Constify mptcp_pm_ops
1375648  [net] tcp: fix syn cookied MPTCP request socket leak


     our repo (by: Davide Caratti, Geliang Tang, Matthieu Baerts, Paolo 
Abeni):

1377493  [net-next] selftests: mptcp: interpret \n as a new line

1377396  [net] net: mptcp: make DACK4/DACK8 usage consistent among all 
subflows

1376740  [net-next] Squash-to: "mptcp: refactor shutdown and close"

1376684  [net-next] Squash-to: mptcp: move page frag allocation in 
mptcp_sendm...

1376301  [net-next] mptcp: ADD_ADDRs with echo bit are smaller

1375632  [v3,13/13] mptcp: send explicit ack on delayed ack_seq incr
1375633  [v3,12/13] mptcp: keep track of advertised windows right edge
1375634  [v3,11/13] mptcp: rework poll+nospace handling
1375631  [v3,10/13] mptcp: try to push pending data on snd una updates
1375630  [v3,09/13] mptcp: move page frag allocation in mptcp_sendmsg()
1375629  [v3,08/13] mptcp: refactor shutdown and close
1375628  [v3,07/13] mptcp: introduce MPTCP snd_nxt
1375626  [v3,06/13] mptcp: add accounting for pending data
1375627  [v3,05/13] mptcp: reduce the arguments of mptcp_sendmsg_frag
1375625  [v3,04/13] mptcp: introduce mptcp_schedule_work
1375624  [v3,03/13] tcp: factor out __tcp_close() helper
1375621  [v3,02/13] mptcp: don't skip needed ack
1375623  [v3,01/13] mptcp: more DATA FIN fixes

1375217  [v2] tcp: fix syn cookied MPTCP request socket leak

1374891  [v5,mptcp-next,2/2] selftests: mptcp: add ADD_ADDR timeout test 
case
1374890  [v5,mptcp-next,1/2] mptcp: add a new sysctl add_addr_timeout



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: Davide Caratti):

1377553  [net] net: mptcp: make DACK4/DACK8 usage consistent among all 
subflows


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

1370700: RFC: [RFC,2/4] tcp: move selected mptcp helpers to tcp.h/mptcp.h
1370701: RFC: [RFC,3/4] mptcp: add mptcp reset option support
1370702: RFC: [RFC,4/4] tcp: parse tcp options contained in reset packets:
     - WIP

1372421: Changes Requested: [v3,mptcp-next,1/2] mptcp: send out 
dedicated ADD_ADDR packet
1372422: Changes Requested: [v3,mptcp-next,2/2] selftests: mptcp: add 
subflow and ADD_ADDR IPv6 test cases:
     - WIP

1374921: Needs Review / ACK: [mptcp-next] mptcp: skip to next candidate 
if subflow has unacked data:
     - Waiting for review
     - *Paolo* will have a look at it later

1375475: New: [mptcp-next] selftests: mptcp: add link failure test case:
     - Depends on the previous patch (mptcp: skip to next candidate if 
subflow has unacked data) not to have other failure in the selftests

1375893: New: [RFC,mptpcp-next] mptcp: add ooo prune support:
     - Waiting for review
     - *Mat* will have a look at it later

1376724: New: [mptcp-next] Squash-to: mptcp: keep track of advertised 
windows right edge:
     - Can be applied: we can enable pr_debug dynamically
     - TODO: *Matth*

1378182: Changes Requested: [net,1/2] mptcp: fix fallback for MP_JOIN 
subflows
1378181: Changes Requested: [net,2/2] mptcp: subflows garbage collection:
     - Will try to send a v2 fixing Mat's comment.


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

     Recently opened (latest from last week: 95)

   99  simult_flows selftest is unstable: remaining sockets in TIME-WAIT 
state [bug]:
       - Could be due to out of window data we have at the end of the 
"export" branch
       - With what Paolo sent, this situation should be less frequent
       - But a final fix (non trivial so that will be for Florian :) ) 
is required

   98  dss_ssn_specified_client packetdrill test fails (timeout) [bug]:
       - seems to be an issue with packetdrill test itself
       - because the data fin are sent first and then the FIN
       - *Davide* has a PR but not fixing this specific issue right now.

   97  null ptr dereference on close timeout under memory pressure [bug] 
@pabeni :
       - introduced by the recent refactor
       - we could still enqueue data to a subflow after the it has been 
closed
       - allocation could fail, send the notification to socket struct 
but will receive a null pointer
       - should be fixed by "mptcp: mp join fallback fixes" series


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

   97  null ptr dereference on close timeout under memory pressure [bug] 
@pabeni

   94  Packetdrill: after a received DATA_FIN, no new packets can be 
treated [bug] @dcaratti :
       - the received packet doesn't match the socket
       - it doesn't match the expected socket

   92  mptcp_close() deletes MPTCP socket before disconnection completes 
[bug] @mjmartineau:
       - should be addressed by the recent sendmsg refactor, right?
       - *Mat* is going to verify this and close it if OK

   91  MPJ: Fallback to TCP if the client receives a 4th ACK without DSS 
options from the server [bug] @pabeni:
       - should be addressed by "mptcp: mp join fallback fixes" series

   85  Packetdrill: multiple timeout reported by the CI [bug] @matttbe:
       - WIP


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

   99  simult_flows selftest is unstable: remaining sockets in TIME-WAIT 
state [bug]
   98  dss_ssn_specified_client packetdrill test fails (timeout) [bug]
   70  [syzkaller] WARNING in mptcp_reset_timer [bug] [syzkaller]
   65  clearing properly the status in listen() [bug]
   56  msk connection state set without msk lock [bug]


     In Progress (opened and assigned)

   96  Python: add support for IPPROTO_MPTCP [enhancement] @matttbe :
       - WIP

   76  [gs]etsockopt per subflow: BPF [enhancement] @matttbe :
       - WIP

   55  ADD_ADDR: IPv6 support [enhancement] @geliangtang
   54  ADD_ADDR: ports support [enhancement] @geliangtang
   43  [syzkaller] Change syzkaller to exercise MPTCP inet_diag 
interface [enhancement] [syzkaller] @cpaasch


     Recently closed (since last week)

   95  Additional random values are set with ADD_ADDR+echo bit packets 
[bug] @matttbe
   89  kmemleak in cookie_tcp_reqsk_alloc (mptcp_join.sh) [bug] @pabeni



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):
         - *@Christoph* Is it running on top of latest export branch 
with sendmsg refactor?
         - There was one report from syzbot but already fixed by Paolo

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

     - news about Intel's kbuild? (Mat):
         - got reports but still difficult to know what's wrong: we only 
have "Test X has failed"
         - Mat will check how to get more details

     - packetdrill (Davide):
         - echo-bit explicit?:
             - not mandatory, but it can be modified

     - CI (Matth):
         - could be good to continue to run on a slow server but also a 
non laggy one



selftests:
     - OK to depend on IPTables?:
         - we should at least skip the test if the tool is not available
         - and probably better to switch to nftables
         - bpftables has been deprecated, we can no longer be cool and 
do it with bpf :'(
         - TODO: *Matth*



Next meeting:
     - We propose to have the next meeting on Thursday, the 15th of October.
     - Usual time: 15:00 UTC (8am PDT, 5pm CEST, 11pm UTC+8)
     - Still open to everyone!
     - https://annuel2.framapad.org/p/mptcp_upstreaming_20201015



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