Re: [PATCH] x86/mce: correct cpu_missing reporting in mce_timed_out

[Borislav Petkov]

On Thu, Nov 04, 2021 at 03:44:31PM +0800, Zhaolong Zhang wrote:
> set cpu_missing before mce_panic() so that it prints correct msg.
> 
> Signed-off-by: Zhaolong Zhang <zhangzl2013@xxxxxxx>
> ---
>  arch/x86/kernel/cpu/mce/core.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/x86/kernel/cpu/mce/core.c b/arch/x86/kernel/cpu/mce/core.c
> index 50a3e455cded..ccefe131ab55 100644
> --- a/arch/x86/kernel/cpu/mce/core.c
> +++ b/arch/x86/kernel/cpu/mce/core.c
> @@ -903,13 +903,13 @@ static int mce_timed_out(u64 *t, const char *msg)
>  	if (!mca_cfg.monarch_timeout)
>  		goto out;
>  	if ((s64)*t < SPINUNIT) {
> +		cpu_missing = 1;
>  		if (mca_cfg.tolerant <= 1) {
>  			if (cpumask_and(&mce_missing_cpus, cpu_online_mask, &mce_missing_cpus))
>  				pr_emerg("CPUs not responding to MCE broadcast (may include false positives): %*pbl\n",
>  					 cpumask_pr_args(&mce_missing_cpus));
>  			mce_panic(msg, NULL, NULL);
>  		}
> -		cpu_missing = 1;
>  		return 1;
>  	}
>  	*t -= SPINUNIT;
> -- 

Frankly, we might just as well kill that cpu_missing thing because we
already say that some CPUs are not responding.

And that "Some CPUs didn't answer in synchronization" is not really
telling me a whole lot.

Tony, do you see any real need to keep it?

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette