Achim Gratz Stromeko at nexgo.de
Sat Apr 13 16:28:19 UTC 2019

Udo van den Heuvel via devel writes:
> On 13-04-19 15:29, Achim Gratz via devel wrote:
>> Udo van den Heuvel via devel writes:
>>> Fedora 29, kernel.org 4.19.30. (my compile)
>> Is this a "tickless" kernel perhaps?  If so, try "nohz=off" on the
>> kernel command line in the boot manager and see if that helps.
> # grep HZ .config|grep -v ^#
> CONFIG_HZ_250=y
> tickless does not go with PPS support...

How about you boot with "nohz=off"?

> My kernel is from kernel.org and compiled by me.

What keeps you from using a newer one, then?

> I need faster memory for later BIOS. Perhaps later this year.

That's the first time I ever hear about a BIOS update requiring a higher
memory speed (earlier BIOS versions not supporting _hugher_ memory
speeds, yes -- but not the other way around).

> !?
> What /was/ the problem in that case?

The same that you still see: the system clock gets stopped and two
consecutive reads of the clock read the same (or maybe time even goes
backwards when looked at with higher resolution).

> With tsc there was no real sync to pps.
> With hpet we do have sync to pps.

The only thing switching to HPET does is reducing clock resolution _a
lot_, so if your clock is funky with TSC, it still is with HPET.  It has
more chance of averaging out over the longer interval between clock
ticks, that's all.

+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

SD adaptation for Waldorf rackAttack V1.04R1:

More information about the devel mailing list