Re: [PATCH] x86,tlb: update mm_cpumask lazily

[Dave Hansen]

On 11/8/24 11:31, Rik van Riel wrote:
>  		/* Can only happen on remote CPUs */
> -		if (f->mm && f->mm != loaded_mm)
> +		if (f->mm && f->mm != loaded_mm) {
> +			cpumask_clear_cpu(raw_smp_processor_id(), mm_cpumask(f->mm));
>  			return;
> +		}

We could also stick a new flush event type in here that says it received
an IPI for a non-loaded mm to see how common those are.