Re: [PATCH] x86/cpu: Add INTEL_LUNARLAKE_M to X86_BUG_MONITOR

[Thomas Gleixner <tglx@linutronix.de>]

On Fri, Nov 08 2024 at 08:49, Len Brown wrote:
> diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
> index e7656cbef68d..aa63f5f780a0 100644
> --- a/arch/x86/kernel/cpu/intel.c
> +++ b/arch/x86/kernel/cpu/intel.c
> @@ -586,7 +586,8 @@ static void init_intel(struct cpuinfo_x86 *c)
>  	     c->x86_vfm == INTEL_WESTMERE_EX))
>  		set_cpu_bug(c, X86_BUG_CLFLUSH_MONITOR);
>  
> -	if (boot_cpu_has(X86_FEATURE_MWAIT) && c->x86_vfm == INTEL_ATOM_GOLDMONT)
> +	if (boot_cpu_has(X86_FEATURE_MWAIT) &&
> +			(c->x86_vfm == INTEL_ATOM_GOLDMONT || c->x86_vfm == INTEL_LUNARLAKE_M))

This indentation is bogus.

>  		set_cpu_bug(c, X86_BUG_MONITOR);

> diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
> index 766f092dab80..910cb2d72c13 100644
> --- a/arch/x86/kernel/smpboot.c
> +++ b/arch/x86/kernel/smpboot.c
> @@ -1377,6 +1377,9 @@ void smp_kick_mwait_play_dead(void)
>  		for (i = 0; READ_ONCE(md->status) != newstate && i < 1000; i++) {
>  			/* Bring it out of mwait */
>  			WRITE_ONCE(md->control, newstate);
> +			/* If MONITOR unreliable, send IPI */
> +			if (boot_cpu_has_bug(X86_BUG_MONITOR))
> +				__apic_send_IPI(cpu, RESCHEDULE_VECTOR);

How is this supposed to work?

The local APIC of the offline CPU is shut down and only responds to
INIT, NMI, SMI, and SIPI.

Even if the APIC would react to the IPI, then the offline CPU would not
notice as is has interrupts disabled when it reaches mwait_play_dead().

Seriously?

Thanks,

        tglx