Re: [LKP] Re: [x86/mce] 1de08dccd3: will-it-scale.per_process_ops -14.1% regression

From: Feng Tang
Date: Tue Aug 25 2020 - 21:46:04 EST


On Wed, Aug 26, 2020 at 12:44:37AM +0800, Luck, Tony wrote:
> > These 2 variables are accessed in 2 hot call stacks (for this 288 CPU
> > Xeon Phi platform):
>
> This might be the key element of "weirdness" for this system. It
> has 288 CPUs ... cache alignment problems are often not too bad
> on "small" systems. The as you scale up to bigger machines you
> suddenly hit some critical point and performance drops dramatically.
>
> It's good that you are picking up tips on how to bisect these and diagnose
> the underlying problem. Number of cores is going to keep increasing, so
> we will keep finding new issues like this.

Yes, now we have one more bullet for shooting this kind of strange
performance changes :)

Thanks,
Feng

> -Tony