| Message ID | 20220504215408.349318-1-mathew.j.martineau@linux.intel.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | mptcp: Improve MPTCP-level window tracking | expand |
On Wed, 4 May 2022, Mat Martineau wrote: > This series improves MPTCP receive window compliance with RFC 8684 and > helps increase throughput on high-speed links. Note that patch 3 makes a > change in tcp_output.c > > For the details, Paolo says: > > I've been chasing bad/unstable performance with multiple subflows > on very high speed links. > > It looks like the root cause is due to the current mptcp-level > congestion window handling. There are apparently a few different > sub-issues: > > - the rcv_wnd is not effectively shared on the tx side, as each > subflow takes in account only the value received by the underlaying > TCP connection. This is addressed in patch 1/4 I missed a couple of small edits - this should be "patch 1/5"... > > - The mptcp-level offered wnd right edge is currently allowed to shrink. > Reading section 3.3.4.: > > """ > The receive window is relative to the DATA_ACK. As in TCP, a > receiver MUST NOT shrink the right edge of the receive window (i.e., > DATA_ACK + receive window). The receiver will use the data sequence > number to tell if a packet should be accepted at the connection > level. > """ > > I read the above as we need to reflect window right-edge tracking > on the wire, see patch 4/5. > > - The offered window right edge tracking can happen concurrently on > multiple subflows, but there is no mutex protection. We need an > additional atomic operation - still patch 3/4 ...and this "patch 4/5". - Mat > > This series additionally bumps a few new MIBs to track all the above > (ensure/observe that the suspected races actually take place). > > I could not access again the host where the issue was so > noticeable, still in the current setup the tput changes from > [6-18] Gbps to 19Gbps very stable. > > > Paolo Abeni (5): > mptcp: really share subflow snd_wnd > mptcp: add mib for xmit window sharing > tcp: allow MPTCP to update the announced window > mptcp: never shrink offered window > mptcp: add more offered MIBs counter > > include/net/mptcp.h | 2 +- > net/ipv4/tcp_output.c | 14 ++++++----- > net/mptcp/mib.c | 4 +++ > net/mptcp/mib.h | 6 +++++ > net/mptcp/options.c | 58 +++++++++++++++++++++++++++++++++++++------ > net/mptcp/protocol.c | 32 +++++++++++++++--------- > net/mptcp/protocol.h | 2 +- > 7 files changed, 90 insertions(+), 28 deletions(-) > > > base-commit: a37f37a2e7f5ea3ae2a1278f552aa21a8e32c221 > -- > 2.36.0 > > -- Mat Martineau Intel
Hello: This series was applied to netdev/net-next.git (master) by Jakub Kicinski <kuba@kernel.org>: On Wed, 4 May 2022 14:54:03 -0700 you wrote: > This series improves MPTCP receive window compliance with RFC 8684 and > helps increase throughput on high-speed links. Note that patch 3 makes a > change in tcp_output.c > > For the details, Paolo says: > > I've been chasing bad/unstable performance with multiple subflows > on very high speed links. > > [...] Here is the summary with links: - [net-next,1/5] mptcp: really share subflow snd_wnd https://git.kernel.org/netdev/net-next/c/b713d0067574 - [net-next,2/5] mptcp: add mib for xmit window sharing https://git.kernel.org/netdev/net-next/c/92be2f522777 - [net-next,3/5] tcp: allow MPTCP to update the announced window https://git.kernel.org/netdev/net-next/c/ea66758c1795 - [net-next,4/5] mptcp: never shrink offered window https://git.kernel.org/netdev/net-next/c/f3589be0c420 - [net-next,5/5] mptcp: add more offered MIBs counter https://git.kernel.org/netdev/net-next/c/38acb6260f60 You are awesome, thank you!