Linux RDMA and InfiniBand development
 help / color / mirror / Atom feed
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 v3 0/2] net/rds: scope-aware sendmsg connection cache
Date: Sat,  3 Oct 2026 09:35:24 -0700	[thread overview]
Message-ID: <20261003163526.250805-1-achender@kernel.org> (raw)

Hi all,

This is v3 of the sendmsg connection-cache scope fix (v1 at [1], v2 at
[2]), now two patches.

  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
  now takes its scope from the connected peer's field, and rds_bind()'s
  connected-socket check compares against it.

  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().

Changes since v2 [2]:
 - The v2 "scope 0 matches anything" wildcard is gone: it sent a
   connected socket's traffic out whichever interface the socket last
   used explicitly (review of v2).  Instead the destination-less send
   carries the connected peer's scope (new patch 1), and the comparison
   is exact.
 - Rebased onto current net.

Changes since v1 [1]:
 - A request without a scope kept matching the cached connection
   (superseded by the above).

[1] https://lore.kernel.org/netdev/20260921215046.174745-1-achender@kernel.org/
[2] https://lore.kernel.org/netdev/20260927062312.169097-1-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 | 12 ++++++++----
 net/rds/bind.c   |  4 ++--
 net/rds/rds.h    |  2 ++
 net/rds/send.c   |  5 +++--
 4 files changed, 15 insertions(+), 8 deletions(-)

-- 
2.25.1


             reply	other threads:[~2026-10-03 16:35 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-03 16:35 Allison Henderson [this message]
2026-10-03 16:35 ` [PATCH net v3 1/2] net/rds: keep the connected peer's scope id apart from the bound one Allison Henderson
2026-10-04 16:35   ` sashiko-bot
2026-10-04 16:41   ` netdev-bot+sashiko
2026-10-03 16:35 ` [PATCH net v3 2/2] net/rds: include the scope id in the sendmsg connection cache check Allison Henderson
2026-10-04 16:35   ` sashiko-bot
2026-10-04 16:41   ` netdev-bot+sashiko
2026-10-03 16:39 ` [PATCH net v3 0/2] net/rds: scope-aware sendmsg connection cache netdev-bot+sinfo

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=20261003163526.250805-1-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