Re: [RFC PATCH net-next] net: macb: Move delayed work on system_dfl_wq

[Nicolai Buchwitz <nb@tipi-net.de>]

On 15.5.2026 17:27, Marco Crivellari wrote:
> Currently the code enqueue work items using {queue|mod}_delayed_work(),
> using system_wq, which will be deprecated soon and replaced by
> system_percpu_wq.
> 
>    commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and 
> system_dfl_wq")
> 
> The function(s) mentioned earlier, 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.
> 
> 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_wq with system_dfl_wq so that the work may benefit from
> scheduler task placement.
> 
> Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
> ---
>  drivers/net/ethernet/cadence/macb_main.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/ethernet/cadence/macb_main.c 
> b/drivers/net/ethernet/cadence/macb_main.c
> index a12aa21244e8..42328b5ca403 100644
> --- a/drivers/net/ethernet/cadence/macb_main.c
> +++ b/drivers/net/ethernet/cadence/macb_main.c
> @@ -672,7 +672,7 @@ static void macb_tx_lpi_work_fn(struct work_struct 
> *work)
>  static void macb_tx_lpi_schedule(struct macb *bp)
>  {
>  	if (bp->eee_active)
> -		mod_delayed_work(system_wq, &bp->tx_lpi_work,
> +		mod_delayed_work(system_dfl_wq, &bp->tx_lpi_work,
>  				 usecs_to_jiffies(bp->tx_lpi_timer));
>  }
> 
> @@ -724,7 +724,7 @@ static int macb_mac_enable_tx_lpi(struct 
> phylink_config *config, u32 timer,
>  	/* Defer initial LPI entry by 1 second after link-up per
>  	 * IEEE 802.3az section 22.7a.
>  	 */
> -	mod_delayed_work(system_wq, &bp->tx_lpi_work, 
> msecs_to_jiffies(1000));
> +	mod_delayed_work(system_dfl_wq, &bp->tx_lpi_work, 
> msecs_to_jiffies(1000));
> 
>  	return 0;
>  }

Reviewed-by: Nicolai Buchwitz <nb@tipi-net.de>

Thanks
Nicolai