[PATCH net] net/rds: size a connection's path set by the transport it ends up with

[PATCH net] net/rds: size a connection's path set by the transport it ends up with

[Allison Henderson]

__rds_conn_create() computes npaths from the caller's transport before
it decides whether a connection to one of the host's own addresses is
to be handled by the loopback transport instead.  That substitution is
what an RDS/TCP socket sending to a local address gets, and after it
the path init loop still runs for the TCP transport's RDS_MPATH_WORKERS
paths and allocates an ordered workqueue for each, while
rds_loop_conn_alloc() only ever provides transport data for path 0.

rds_conn_destroy() sizes its teardown from c_trans, by then the
loopback transport, so it visits path 0 only - and
rds_conn_path_destroy() would skip the other paths anyway, since it
returns before destroy_workqueue() for a path without transport data.
kfree(c_path) then drops the last pointers to seven workqueues.  That
repeats for every such connection, on every netns teardown or module
unload, and every distinct local destination address is a separate
connection.

Recompute npaths once the transport is final, so that creation and
destruction agree on the set of paths.  The c_path array stays sized
for the caller's transport; the unused entries are freed with it.

Fixes: 4716af3897e9 ("net/rds: Give each connection path its own workqueue")
Assisted-by: Claude-Code:claude-fable-5
Signed-off-by: Allison Henderson <achender@kernel.org>
---
 net/rds/connection.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/net/rds/connection.c b/net/rds/connection.c
index b6c4beb50eaf..c752a8623cfc 100644
--- a/net/rds/connection.c
+++ b/net/rds/connection.c
@@ -276,6 +276,12 @@ static struct rds_connection *__rds_conn_create(struct net *net,
 
 	conn->c_trans = trans;
 
+	/* The transport may just have been swapped for loopback; size the
+	 * set of paths - which is also what rds_conn_destroy() tears down
+	 * again - by the transport the connection actually uses.
+	 */
+	npaths = (trans->t_mp_capable ? RDS_MPATH_WORKERS : 1);
+
 	init_waitqueue_head(&conn->c_hs_waitq);
 	for (i = 0; i < npaths; i++) {
 		__rds_conn_path_init(conn, &conn->c_path[i],
-- 
2.25.1

Re: [PATCH net] net/rds: size a connection's path set by the transport it ends up with

[sashiko-bot]

> __rds_conn_create() computes npaths from the caller's transport before
> it decides whether a connection to one of the host's own addresses is
> to be handled by the loopback transport instead.  That substitution is
> what an RDS/TCP socket sending to a local address gets, and after it
> the path init loop still runs for the TCP transport's RDS_MPATH_WORKERS
> paths and allocates an ordered workqueue for each, while
> rds_loop_conn_alloc() only ever provides transport data for path 0.
> 
> rds_conn_destroy() sizes its teardown from c_trans, by then the
> loopback transport, so it visits path 0 only - and
> rds_conn_path_destroy() would skip the other paths anyway, since it
> returns before destroy_workqueue() for a path without transport data.
> kfree(c_path) then drops the last pointers to seven workqueues.  That
> repeats for every such connection, on every netns teardown or module
> unload, and every distinct local destination address is a separate
> connection.
> 
> Recompute npaths once the transport is final, so that creation and
> destruction agree on the set of paths.  The c_path array stays sized
> for the caller's transport; the unused entries are freed with it.
> 
> Fixes: 4716af3897e9 ("net/rds: Give each connection path its own workqueue")
> Assisted-by: Claude-Code:claude-fable-5
> Signed-off-by: Allison Henderson <achender@kernel.org>

Sashiko has reviewed this patch and found no issues. It looks great!

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260921215027.174657-1-achender@kernel.org?part=1

Re: [PATCH net] net/rds: size a connection's path set by the transport it ends up with

[patchwork-bot+netdevbpf]

Hello:

This patch was applied to netdev/net.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Mon, 21 Sep 2026 14:50:27 -0700 you wrote:
> __rds_conn_create() computes npaths from the caller's transport before
> it decides whether a connection to one of the host's own addresses is
> to be handled by the loopback transport instead.  That substitution is
> what an RDS/TCP socket sending to a local address gets, and after it
> the path init loop still runs for the TCP transport's RDS_MPATH_WORKERS
> paths and allocates an ordered workqueue for each, while
> rds_loop_conn_alloc() only ever provides transport data for path 0.
> 
> [...]

Here is the summary with links:
  - [net] net/rds: size a connection's path set by the transport it ends up with
    https://git.kernel.org/netdev/net/c/ba3d1f480c7a

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html