Re: Benchmarks: FreeBSD 13 vs. NetBSD 9.2 vs. OpenBSD 7 vs. DragonFlyBSD 6 vs. Linux

Hi there, sorry for the ignorance, but which are the best settings for 
sysctl?

I guess, it depends on the workload that one is executing, right? But 
based on the emails, I understand some basic ones can be applied to help 
with general applications, am I right?

If that's the case, can we attempt to list them? I am happy to create a 
list, document and categorize them based on workloads, I believe it will 
be a good starting point for normal users (like me).

I do have a small list of sysctls that I usually apply, but they are all 
mostly related to networking and particular to our client's setup.

Best regards.

Santi


On 12/11/21 12:20, dmilith . wrote:
> Maybe FreeBSD wouldn't win but results would be much closer. It's also
> known that default compiler options for ports aren't best. Default
> sysctl.conf default settings aren't best. They could at least compile all
> the benchmarks with the same compiler features. So -ffast-math, -flto and
> -O3 + code hardening features, for all tested systems. It's also
> professional to mention the compiler used (I recall that the previous
> Phroenix "benchmark" was done using GCC on FreeBSD which I'll not even
> comment).
>
> I could go on with all mistakes made in this "benchmark", but well, I know
> - benchmarking is hard.
>
>
> On Sat, 11 Dec 2021 at 12:01, Miroslav Lachman <000.fbsd@quip.cz> wrote:
>
>> On 11/12/2021 11:17, dmilith . wrote:
>>> 1. Where are compiler options for BSDs?
>>> 2. Why they compare -O2 to -O3 code in some benchmarks? Why they enable
>>> fast math in some, and disable it for others?
>>> 3. Why they don't mention powerd setup for FreeBSD? By default it may use
>>> slowest CPU mode. Did they even load cpufreq kernel module?
>>> 4. Did they even care about default FreeBSD mitigations (via sysctl)
>>> enabled, or it's only valid for Linuxes? ;)
>>> 5. What happened to security and environment details of BSDs?
>>>
>>> It's kinda known that guys from Phroenix lack basic knowledge of how to
>> do
>>> proper performance testing and lack basic knowledge about BSD systems.
>>> Nothing new. Would take these results with a grain of salt.
>> It is very simple - they are comparing OSes with setting they are
>> shipped. Average users don't know about tuning so the benchmark is
>> reflecting what many average users get.
>> And to be honest - I don't think FreeBSD will win even with everything
>> best tuned.
>>
>> Kind regards
>> Miroslav Lachman
>>
>