Re: [x86/entry] 8c5045f89b: will-it-scale.per_process_ops -12.9% regression

[Kees Cook]

On Fri, Mar 27, 2020 at 03:40:02PM +0800, kernel test robot wrote:
> Greeting,
> 
> FYI, we noticed a -12.9% regression of will-it-scale.per_process_ops due to commit:
> 
> 
> commit: 8c5045f89b181a4412b698cafc9dc257192a3a65 ("x86/entry: Enable random_kstack_offset support")
> https://git.kernel.org/cgit/linux/kernel/git/kees/linux.git kspp/stack/offset/agnostic/v1
> 
> in testcase: will-it-scale
> on test machine: 288 threads Intel(R) Xeon Phi(TM) CPU 7295 @ 1.50GHz with 80G memory
> with following parameters:
> 
> 	nr_task: 100%
> 	mode: process
> 	test: lseek1
> 	cpufreq_governor: performance
> 	ucode: 0x11
> 
> test-description: Will It Scale takes a testcase and runs it from 1 through to n parallel copies to see if the testcase will scale. It builds both a process and threads based test in order to see any differences between the two.
> test-url: https://github.com/antonblanchard/will-it-scale

Whoa. I'll take a closer look. I fail to understand how this could be
happening. In this CONFIG the changes are a NOP. And that would be
true for both process and thread. O_o

-- 
Kees Cook