Matthieu Baerts

in-kernel PM: store received `ADD_ADDR`

1

Currently, upon the reception of an `ADD_ADDR`, new subflows are potentially created (depending on the limits), and accepted `ADD_ADDR` are somehow stored in the 'conn_list'. If we want to browse...

It looks like the subflow counter is not updated when the other end initiates the removal of a subflow, e.g. after having received a RM_ADDR: ```diff diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh...

MPTCP and KTLS support

10

Currently, an MPTCP socket cannot support other TCP ULP extensions. This includes KTLS. KTLS is supported in GnuTLS and OpenSSL libraries. It is then tempting to use it without too...

selftests: `mptcp_join`: tests `Infinite map` & `MP_FAIL MP_RST` are unstable

2

It looks like we didn't have this issue recently on our side, but the NetDev CI got it a few times, e.g. ``` # 104 Infinite map # Info: Test...

BPF: `setsockopt` on an MPTCP socket: check support

1

We have recommended to use BPF to set socket option per subflow (see #76), on TCP socket then. But it looks like we didn't check if we can correctly do...

[6.1] selftests: `mptcp_connect.sh` failures with `copyfd_io_poll: poll timed out` errors

3

nkambo from LKFT reported that `mptcp_connect.sh` subtests are, e.g. ``` # INFO: with peek mode: saveWithPeek # ns1 MPTCP -> ns1 (10.0.1.1:10042 ) MPTCP (duration 1024ms) [ OK ] #...

See Eric's commit: 151c9c724d05d5b0dd8acd3e11cb69ef1f2dbada https://lore.kernel.org/r/[email protected] We can then remove: ```c /* kernel sockets do not by default acquire net ref, but TCP timer * needs it. * Update ns_tracker to...

Many microservices packaged in a container uses AlpineLinux as a base. It would be helpful to have mptcpd available there, e.g. to easily run 'mptcpize'. https://wiki.alpinelinux.org/wiki/Creating_an_Alpine_package

**Describe the bug** It looks like ELL is going to break the API around `l_netlink_send()`: [commit](https://git.kernel.org/pub/scm/libs/ell/ell.git/commit/?id=883bc6a90b4a9c654fdf2540f02447a7ca567e04) **To Reproduce** Steps to reproduce the behavior: 1. Run the CI, _[ELL master /...

This has been covered for the `msk->fully_established` field,, but it looks like it is not the case for the subflow one: msk: ``` net/mptcp/pm_netlink.c: if (!READ_ONCE(msk->fully_established) || net/mptcp/protocol.c: WRITE_ONCE(msk->fully_established, false);...