# Re: \[fs\] a0918006f9: netperf.Throughput\_tps -11.6% regression

- From: Yin Fengwei \<hidden\>
- Date: 2021-11-11 03:30:40 
- List: linux-security-module
- Page: https://ratatoskr.run/linux-security-module/2021/11/217957
- Thread: https://ratatoskr.run/linux-api/2021/10/215793/t.md

```text

On 11/10/2021 4:52 PM, Mickaël Salaün wrote:
>>> ---------------- ---------------------------
>>>          %stddev     %change         %stddev
>>>              \          |                \
>>>     555600            -0.1%     555305        netperf.Throughput_total_tps
>>>      34725            -0.1%      34706        netperf.Throughput_tps
>>>
>>>
>>> Fengwei also helped review these results and commented:
>>> I suppose these three CPUs have different cache policy. It also could be
>>> related with netperf throughput testing.
>> Does moving the syscall implementation somewhere else change things?
>> That's a _huge_ performance change for something that isn't even called.
>> What's going on here?
> This regression doesn't make sense. I guess this is the result of a
> flaky netperf test, maybe because the test machine was overloaded at
> that time.

I agree the test result looks strange. But I don't think the test machine
or test methodology has issue. It's not possible the test box is overloaded
when test case is running. We did test several times (> 12 times) on different
days. Thanks.


Regards
Yin, Fengwei
```