Re: [PATCH] thp: reduce usage of huge zero page's atomic counter

[Anshuman Khandual]

On 08/30/2016 09:09 AM, Andrew Morton wrote:
> On Tue, 30 Aug 2016 11:09:15 +0800 Aaron Lu <aaron.lu@xxxxxxxxx> wrote:
> 
>>>> Case used for test on Haswell EP:
>>>> usemem -n 72 --readonly -j 0x200000 100G
>>>> Which spawns 72 processes and each will mmap 100G anonymous space and
>>>> then do read only access to that space sequentially with a step of 2MB.
>>>>
>>>> perf report for base commit:
>>>>     54.03%  usemem   [kernel.kallsyms]   [k] get_huge_zero_page
>>>> perf report for this commit:
>>>>      0.11%  usemem   [kernel.kallsyms]   [k] mm_get_huge_zero_page
>>>
>>> Does this mean that overall usemem runtime halved?
>>
>> Sorry for the confusion, the above line is extracted from perf report.
>> It shows the percent of CPU cycles executed in a specific function.
>>
>> The above two perf lines are used to show get_huge_zero_page doesn't
>> consume that much CPU cycles after applying the patch.
>>
>>>
>>> Do we have any numbers for something which is more real-wordly?
>>
>> Unfortunately, no real world numbers.
>>
>> We think the global atomic counter could be an issue for performance
>> so I'm trying to solve the problem.
> 
> So, umm, we don't actually know if the patch is useful to anyone?

On a POWER system it improves the CPU consumption of the above mentioned
function a little bit. Dont think its going to improve actual throughput
of the workload substantially.

0.07%  usemem  [kernel.vmlinux]  [k] mm_get_huge_zero_page

to

0.01%  usemem  [kernel.vmlinux]  [k] mm_get_huge_zero_page