Re: [Patch v4 12/16] sched/cache: Fix race condition during sched domain rebuild

[Chen, Yu C]

On 5/18/2026 10:34 PM, Peter Zijlstra wrote:
> On Wed, May 13, 2026 at 01:39:23PM -0700, Tim Chen wrote:
> >   /*
> >    * Enable/disable cache aware scheduling according to
> >    * user input and the presence of hardware support.
> > + * Expected to be protected by cpus_read_lock() and
> > + * sched_domains_mutex_lock()
> >    */
> > -static void sched_cache_active_set(bool locked)
> > +static void _sched_cache_active_set(void)
>
> Like so?

Yes, I think this is OK. One case is that the _sched_cache_active_set()
might be called via sched_init_smp() -> sched_init_domains() -> 
build_sched_domains()
where only sched_domains_mutex  was hold. But since it is during boot
up thus system_state < SYSTEM_RUNNING and lockdep_assert_cpus_held() will
not trigger the assert.

thanks,
Chenyu


> --- a/kernel/sched/topology.c
> +++ b/kernel/sched/topology.c
> @@ -920,11 +920,12 @@ static bool alloc_sd_llc(const struct cp
>   /*
>    * Enable/disable cache aware scheduling according to
>    * user input and the presence of hardware support.
> - * Expected to be protected by cpus_read_lock() and
> - * sched_domains_mutex_lock()
>    */
>   static void _sched_cache_active_set(void)
>   {
> +	lockdep_assert_cpus_held();
> +	lockdep_assert_held(&sched_domains_mutex);
> +
>   	/* hardware does not support */
>   	if (!static_branch_likely(&sched_cache_present)) {
>   		static_branch_disable_cpuslocked(&sched_cache_active);