Re: [PATCH 2/2] sched/fair: Scale wakeup granularity relative to nr_running
From: Peter Zijlstra
Date: Tue Oct 05 2021 - 06:38:24 EST
On Thu, Sep 23, 2021 at 08:24:03AM -0400, Phil Auld wrote:
> It's capped at 8 cpus, which is pretty easy to reach these days, so the
> values don't get too large. That scaling is almost a no-op these days.
https://lkml.kernel.org/r/YVwdrh5pg0zSv2/b@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Ooh, hey, we already fixed that :-)
So the reasoning there is that if the values get too big, interactiviy
get *really* bad, but if you go from say 1 to 4 CPUs, interactivity can
improve due to being able to run on other CPUs.
At 8 CPUs we end up at 6ms*4=24ms, which is already pretty terrible.