[PATCH v2] xprtrdma: Move long delayed work on system_dfl_long_wq

[PATCH v2] xprtrdma: Move long delayed work on system_dfl_long_wq

[Marco Crivellari]

Currently the code enqueue work items using {queue|mod}_delayed_work(),
using system_long_wq. This workqueue should be used when long works are
expected and it is a per-cpu workqueue.

The function(s) end up calling __queue_delayed_work(), which set a global
timer that could fire anywhere, enqueuing the work where the timer fired.

Unbound works could benefit from scheduler task placement, to optimize
performance and power consumption. Long work shouldn't stick to a single
CPU.

Recently, a new unbound workqueue specific for long running work has
been added:

    c116737e972e ("workqueue: Add system_dfl_long_wq for long unbound works")

Since the workqueue work doesn't rely on per-cpu variables, there is no
obvious reason that justify the use of a per-cpu workqueue. So change
system_long_wq with system_dfl_long_wq so that the work may benefit from
scheduler task placement.

Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
---
Changes in v2:
- Commit log improvements

- Rebase on v7.1-rc2

Link to v1: https://lore.kernel.org/all/20260430085412.96961-1-marco.crivellari@suse.com/

 net/sunrpc/xprtrdma/transport.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
index 61706df5e485..1a54993f7ffb 100644
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -484,7 +484,8 @@ xprt_rdma_connect(struct rpc_xprt *xprt, struct rpc_task *task)
 		xprt_reconnect_backoff(xprt, RPCRDMA_INIT_REEST_TO);
 	}
 	trace_xprtrdma_op_connect(r_xprt, delay);
-	queue_delayed_work(system_long_wq, &r_xprt->rx_connect_worker, delay);
+	queue_delayed_work(system_dfl_long_wq, &r_xprt->rx_connect_worker,
+			   delay);
 }
 
 /**
-- 
2.53.0

Re: [PATCH v2] xprtrdma: Move long delayed work on system_dfl_long_wq

[Chuck Lever]

On Thu, May 7, 2026, at 3:01 PM, Marco Crivellari wrote:
> Currently the code enqueue work items using {queue|mod}_delayed_work(),
> using system_long_wq. This workqueue should be used when long works are
> expected and it is a per-cpu workqueue.
>
> The function(s) end up calling __queue_delayed_work(), which set a global
> timer that could fire anywhere, enqueuing the work where the timer fired.
>
> Unbound works could benefit from scheduler task placement, to optimize
> performance and power consumption. Long work shouldn't stick to a single
> CPU.
>
> Recently, a new unbound workqueue specific for long running work has
> been added:
>
>     c116737e972e ("workqueue: Add system_dfl_long_wq for long unbound works")
>
> Since the workqueue work doesn't rely on per-cpu variables, there is no
> obvious reason that justify the use of a per-cpu workqueue. So change
> system_long_wq with system_dfl_long_wq so that the work may benefit from
> scheduler task placement.
>
> Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
> ---
> Changes in v2:
> - Commit log improvements
>
> - Rebase on v7.1-rc2
>
> Link to v1: 
> https://lore.kernel.org/all/20260430085412.96961-1-marco.crivellari@suse.com/

Reviewed-by: Chuck Lever <chuck.lever@oracle.com>


-- 
Chuck Lever