The rest of the 6.3 merge window [LWN.net]

The rest of the 6.3 merge window

Posted Mar 7, 2023 7:06 UTC (Tue) by donald.buczek (subscriber, #112892)
Parent article: The rest of the 6.3 merge window

> The kernel now can feature a built-in Dhrystone benchmark test.
> https://git.kernel.org/linus/d5528cc16893
> > author Geert Uytterhoeven <geert+renesas@glider.be> 2022-12-08 15:31:28 +0100
> > committer Andrew Morton <akpm@linux-foundation.org> 2023-02-02 22:50:01 -0800
> > [...]
> > + * Author: Reinhold P. Weicker
> > + * Siemens AG, AUT E 51
> > + * Postfach 3220
> > + * 8520 Erlangen
> > + * Germany (West)
> > + * Phone: [+49]-9131-7-20330
> > + * (8-17 Central European Time)
> > + * Usenet: ..!mcsun!unido!estevax!weicker

Nice bit of nostalgia to have something with a bang path email address merged in 2023.

to post comments

The rest of the 6.3 merge window

Posted Mar 7, 2023 7:32 UTC (Tue) by mb (subscriber, #50428) [Link] (3 responses)

*checking my calendar*.

Yeah, kind of nice.
The coding style is awful, though. And not very kernel like at all.
I would have expected it to be cleaned up before getting merged.
The kernel is not a museum for historic code dumps.

The rest of the 6.3 merge window

Posted Mar 8, 2023 1:29 UTC (Wed) by edeloget (subscriber, #88392) [Link] (1 responses)

> *checking my calendar*.

> Yeah, kind of nice.
> The coding style is awful, though. And not very kernel like at all.
>I would have expected it to be cleaned up before getting merged.
> The kernel is not a museum for historic code dumps.

I guess the style has been kept in order to make sure that the code was doing exactly what it says it does, including the "I need to make sure that the output can be compared to all other DMIPS outputs out there". I haven't looked the whole patch, but this may even come with special compiler options in order to make sure the compiler is not doing some nasty optimization that would lead to non-comparable results.

The rest of the 6.3 merge window

Posted Mar 8, 2023 9:13 UTC (Wed) by geert (subscriber, #98403) [Link]

Indeed, I only changed whitespace and basic formatting to please scripts/checkpatch.pl, to avoid impact on the measurements:
https://git.kernel.org/pub/scm/linux/kernel/git/geert/ren...
In theory, a synthetic benchmark can be optimized away entirely by a sufficiently smart compiler. That's also why the benchmark is split in two source files.

The figures cannot be compared accurately, as they depend heavily on compiler version and compiler options (some dictated by the kernel build system). They do work fine for my intended usage (verifying CPU core clock speeds), as DMIPS/MHz is reliable.

Typically, on e.g. arm32, I get 25% less DMIPS with the in-kernel version than with the userspace version (https://github.com/qris/dhrystone-deb.git).
Even with the latter, I saw differences up to 19% using gcc-4.9/6.3/7.4/8.3/9.3/10.2 (6.3 was worst, 10.2 was best).

The rest of the 6.3 merge window

Posted Mar 9, 2023 11:22 UTC (Thu) by Karellen (subscriber, #67644) [Link]

*checking my calendar*

Thu 10782 Sep 11:19:37 UTC 1993

Seems legit :-)

The rest of the 6.3 merge window

Posted Mar 7, 2023 8:46 UTC (Tue) by Sesse (subscriber, #53779) [Link] (2 responses)

I do feel like including Dhrystone is already nostalgia in itself! Now not only BogoMIPS, but actual, official MIPS…

The rest of the 6.3 merge window

Posted Mar 7, 2023 9:14 UTC (Tue) by geert (subscriber, #98403) [Link]

Now we just need someone to try it on linux-vax ;-)

The rest of the 6.3 merge window

Posted Mar 13, 2023 7:44 UTC (Mon) by anton (subscriber, #25547) [Link]

BogoMIPS are real MIPS (although for a pointless loop), while AFAIK Dhrystone MIPS are based on the idea that a VAX 11/780 executes the Dhrystone benchmark at 1 (VAX) MIPS; however, from what I read, the VAX 11/780 typically ran at 0.5MIPS (about 10 cycles/instruction at 5MHz clock rate), so DMIPS were not real MIPS even on the VAX. These days we can determine the number of instructions executed by Dhrystone easily with perf stat -e instructions.

The rest of the 6.3 merge window

Posted Mar 7, 2023 13:30 UTC (Tue) by amw (subscriber, #29081) [Link] (4 responses)

Not to mention the "Germany (West)".

The rest of the 6.3 merge window

Posted Mar 7, 2023 13:36 UTC (Tue) by timon (subscriber, #152974) [Link] (3 responses)

... and, more subtly, the 4-digit postal code.

The rest of the 6.3 merge window

Posted Mar 7, 2023 13:43 UTC (Tue) by amw (subscriber, #29081) [Link]

I was going to mention that, but then thought better of it.

The rest of the 6.3 merge window

Posted Mar 7, 2023 19:50 UTC (Tue) by geert (subscriber, #98403) [Link]

OMG, Erlangen used to have a MOS Technology Complex Interface Adapter postal code ;-)

The rest of the 6.3 merge window

Posted Mar 8, 2023 0:24 UTC (Wed) by Cyberax (✭ supporter ✭, #52523) [Link]

And now I've spent 2 hours reading about the postal system in Germany.

I blame you.

The rest of the 6.3 merge window

Posted Mar 7, 2023 20:46 UTC (Tue) by MarcB (subscriber, #101804) [Link]

> Nice bit of nostalgia to have something with a bang path email address merged in 2023.

It nicely matches the pre-1993, four-digit German postal code and the "Germany (West)" country designation :-)