From: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
To: netdev@vger.kernel.org
Cc: davem@davemloft.net, kuba@kernel.org, edumazet@google.com,
pabeni@redhat.com, dsahern@kernel.org, horms@kernel.org,
idosch@nvidia.com, kuniyu@amazon.com,
Willem de Bruijn <willemb@google.com>
Subject: [PATCH net-next v2 0/3] ip: improve tcp sock multipath routing
Date: Thu, 24 Apr 2025 10:35:17 -0400 [thread overview]
Message-ID: <20250424143549.669426-1-willemdebruijn.kernel@gmail.com> (raw)
From: Willem de Bruijn <willemb@google.com>
Improve layer 4 multipath hash policy for local tcp connections:
patch 1: Select a source address that matches the nexthop device.
Due to tcp_v4_connect making separate route lookups for saddr
and route, the two can currently be inconsistent.
patch 2: Use all paths when opening multiple local tcp connections to
the same ip address and port.
patch 3: Test the behavior. Extend the fib_tests.sh testsuite with one
opening many connections, and count SYNs on both egress
devices, for packets matching the source address of the dev.
Changelog in the individual patches
Willem de Bruijn (3):
ipv4: prefer multipath nexthop that matches source address
ip: load balance tcp connections to single dst addr and port
selftests/net: test tcp connection load balancing
include/net/flow.h | 1 +
include/net/ip_fib.h | 3 +-
include/net/route.h | 3 +
net/ipv4/fib_semantics.c | 39 +++++---
net/ipv4/route.c | 15 ++-
net/ipv6/route.c | 13 ++-
net/ipv6/tcp_ipv6.c | 2 +
tools/testing/selftests/net/fib_tests.sh | 120 ++++++++++++++++++++++-
tools/testing/selftests/net/lib.sh | 24 +++++
9 files changed, 197 insertions(+), 23 deletions(-)
--
2.49.0.805.g082f7c87e0-goog
next reply other threads:[~2025-04-24 14:35 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-04-24 14:35 Willem de Bruijn [this message]
2025-04-24 14:35 ` [PATCH net-next v2 1/3] ipv4: prefer multipath nexthop that matches source address Willem de Bruijn
2025-04-24 16:47 ` Eric Dumazet
2025-04-25 14:59 ` Ido Schimmel
2025-04-26 15:01 ` Willem de Bruijn
2025-04-27 17:30 ` Ido Schimmel
2025-04-28 16:26 ` Willem de Bruijn
2025-04-29 7:54 ` Ido Schimmel
2025-04-24 14:35 ` [PATCH net-next v2 2/3] ip: load balance tcp connections to single dst addr and port Willem de Bruijn
2025-04-24 16:05 ` David Ahern
2025-04-24 16:51 ` Eric Dumazet
2025-04-25 15:14 ` Ido Schimmel
2025-04-24 14:35 ` [PATCH net-next v2 3/3] selftests/net: test tcp connection load balancing Willem de Bruijn
2025-04-25 15:47 ` Ido Schimmel
2025-04-29 14:40 ` [PATCH net-next v2 0/3] ip: improve tcp sock multipath routing patchwork-bot+netdevbpf
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20250424143549.669426-1-willemdebruijn.kernel@gmail.com \
--to=willemdebruijn.kernel@gmail.com \
--cc=davem@davemloft.net \
--cc=dsahern@kernel.org \
--cc=edumazet@google.com \
--cc=horms@kernel.org \
--cc=idosch@nvidia.com \
--cc=kuba@kernel.org \
--cc=kuniyu@amazon.com \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=willemb@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).