Re: [PATCH] sched/ext: Suppress warning in __this_cpu_write() by disabling preemption

[Peter Zijlstra]

On Wed, Jul 16, 2025 at 05:46:15AM -0700, Breno Leitao wrote:
> __this_cpu_write() emits a warning if used with preemption enabled.
> 
> Function update_locked_rq() might be called with preemption enabled,
> which causes the following warning:
> 
> 	BUG: using __this_cpu_write() in preemptible [00000000] code: scx_layered_6-9/68770
> 
> Disable preemption around the __this_cpu_write() call in
> update_locked_rq() to suppress the warning, without affecting behavior.
> 
> If preemption triggers a  jump to another CPU during the callback it's
> fine, since we would track the rq state on the other CPU with its own
> local variable.
> 
> Suggested-by: Andrea Righi <arighi@xxxxxxxxxx>
> Signed-off-by: Breno Leitao <leitao@xxxxxxxxxx>
> Fixes: 18853ba782bef ("sched_ext: Track currently locked rq")
> Acked-by: Andrea Righi <arighi@xxxxxxxxxx>
> ---
>  kernel/sched/ext.c | 7 +++++++
>  1 file changed, 7 insertions(+)
> 
> diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c
> index b498d867ba210..24fcbd7331f73 100644
> --- a/kernel/sched/ext.c
> +++ b/kernel/sched/ext.c
> @@ -1258,7 +1258,14 @@ static inline void update_locked_rq(struct rq *rq)
>  	 */
>  	if (rq)
>  		lockdep_assert_rq_held(rq);
> +	/*
> +	 * __this_cpu_write() emits a warning when used with preemption enabled.
> +	 * While there's no functional issue if the callback runs on another
> +	 * CPU, we disable preemption here solely to suppress that warning.
> +	 */
> +	preempt_disable();
>  	__this_cpu_write(locked_rq, rq);
> +	preempt_enable();
>  }

This looks dodgy as heck. Why can't we do this update_locked_rq(NULL)
call while still holding rq? Also, I don't seem to have this scx_layered
thing.