From: Allison Henderson <achender@kernel.org>
To: netdev@vger.kernel.org, linux-rdma@vger.kernel.org,
pabeni@redhat.com, edumazet@google.com, kuba@kernel.org,
horms@kernel.org
Cc: achender@kernel.org
Subject: [PATCH net v4 0/2] net/rds: scope-aware sendmsg connection cache
Date: Wed, 7 Oct 2026 20:14:24 -0700 [thread overview]
Message-ID: <20261008031426.1142344-1-achender@kernel.org> (raw)
Hi all,
This is v4 of the sendmsg connection-cache scope fix (v1 at [1], v2 at
[2], v3 at [3]).
Patch 1 gives the connected peer's link-local scope its own field.
rds_connect() used to store it in rs_bound_scope_id, the scope of
the socket's own bound address, which a later bind() to a
non-link-local address resets to 0 and which a later connect()
pollutes for a globally bound socket. A send without a destination,
getpeername() and recvmsg() now take the connected peer's scope from
that field.
Patch 2 compares c_dev_if with the request's scope in the cache-hit
test, so the cache uses the same key as rds_conn_lookup().
Both were found by inspection while reworking this cache for the
connection-lifetime series on net-next; neither has been observed in
the field. The misdelivery in patch 2 needs a socket that sends to
the same link-local address through two interfaces, and the stall in
patch 1 a connect()/bind() order that loses the peer's scope.
Changes since v3 [3]:
- Patch 1: a destination-less send falls back to the bound scope for a
non-link-local peer, so send() and sendto() to the connected peer
compute the same scope (review of v3: a socket bound link-local and
connected to a global peer stalled otherwise); getpeername() and
recvmsg() report the connected peer's scope instead of the bound
scope (review of v3).
- Commit messages say how the issues were found.
Changes since v2 [2]:
- The v2 "scope 0 matches anything" wildcard is gone; the
destination-less send carries the connected peer's scope instead
(new patch 1), and the comparison is exact.
[1] https://lore.kernel.org/netdev/20260921215046.174745-1-achender@kernel.org/
[2] https://lore.kernel.org/netdev/20260927062312.169097-1-achender@kernel.org/
[3] https://lore.kernel.org/netdev/20261003163526.250805-1-achender@kernel.org/
Thank you,
Allison
Allison Henderson (2):
net/rds: keep the connected peer's scope id apart from the bound one
net/rds: include the scope id in the sendmsg connection cache check
net/rds/af_rds.c | 15 +++++++++------
net/rds/bind.c | 4 ++--
net/rds/rds.h | 2 ++
net/rds/recv.c | 3 ++-
net/rds/send.c | 9 +++++++--
5 files changed, 22 insertions(+), 11 deletions(-)
--
2.25.1
next reply other threads:[~2026-10-08 3:14 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-08 3:14 Allison Henderson [this message]
2026-10-08 3:14 ` [PATCH net v4 1/2] net/rds: keep the connected peer's scope id apart from the bound one Allison Henderson
2026-10-08 3:23 ` sashiko-bot
2026-10-09 11:57 ` Simon Horman
2026-10-08 3:14 ` [PATCH net v4 2/2] net/rds: include the scope id in the sendmsg connection cache check Allison Henderson
2026-10-08 3:23 ` sashiko-bot
2026-10-09 11:57 ` Simon Horman
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=20261008031426.1142344-1-achender@kernel.org \
--to=achender@kernel.org \
--cc=edumazet@google.com \
--cc=horms@kernel.org \
--cc=kuba@kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.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).