lkml.org 
[lkml]   [2019]   [Sep]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: [RFC PATCH 1/9] sched,cgroup: Add interface for latency-nice
Date

On Thu, Sep 05, 2019 at 12:13:47 +0100, Qais Yousef wrote...

> On 09/05/19 12:46, Peter Zijlstra wrote:
>> On Thu, Sep 05, 2019 at 10:45:27AM +0100, Patrick Bellasi wrote:
>>
>> > > From just reading the above, I would expect it to have the range
>> > > [-20,19] just like normal nice. Apparently this is not so.
>> >
>> > Regarding the range for the latency-nice values, I guess we have two
>> > options:
>> >
>> > - [-20..19], which makes it similar to priorities
>> > downside: we quite likely end up with a kernel space representation
>> > which does not match the user-space one, e.g. look at
>> > task_struct::prio.
>> >
>> > - [0..1024], which makes it more similar to a "percentage"
>> >
>> > Being latency-nice a new concept, we are not constrained by POSIX and
>> > IMHO the [0..1024] scale is a better fit.
>> >
>> > That will translate into:
>> >
>> > latency-nice=0 : default (current mainline) behaviour, all "biasing"
>> > policies are disabled and we wakeup up as fast as possible
>> >
>> > latency-nice=1024 : maximum niceness, where for example we can imaging
>> > to turn switch a CFS task to be SCHED_IDLE?
>>
>> There's a few things wrong there; I really feel that if we call it nice,
>> it should be like nice. Otherwise we should call it latency-bias and not
>> have the association with nice to confuse people.
>>
>> Secondly; the default should be in the middle of the range. Naturally
>> this would be a signed range like nice [-(x+1),x] for some x. but if you
>> want [0,1024], then the default really should be 512, but personally I
>> like 0 better as a default, in which case we need negative numbers.
>>
>> This is important because we want to be able to bias towards less
>> importance to (tail) latency as well as more importantance to (tail)
>> latency.
>>
>> Specifically, Oracle wants to sacrifice (some) latency for throughput.
>> Facebook OTOH seems to want to sacrifice (some) throughput for latency.
>
> Another use case I'm considering is using latency-nice to prefer an idle CPU if
> latency-nice is set otherwise go for the most energy efficient CPU.
>
> Ie: sacrifice (some) energy for latency.
>
> The way I see interpreting latency-nice here as a binary switch. But maybe we
> can use the range to select what (some) energy to sacrifice mean here. Hmmm.

I see this concept possibly evolving into something more then just a
binary switch. Not yet convinced if it make sense and/or it's possible
but, in principle, I was thinking about these possible usages for CFS
tasks:

- dynamically tune the policy of a task among SCHED_{OTHER,BATCH,IDLE}
depending on crossing certain pre-configured threshold of latency
niceness.

- dynamically bias the vruntime updates we do in place_entity()
depending on the actual latency niceness of a task.

- bias the decisions we take in check_preempt_tick() still depending
on a relative comparison of the current and wakeup task latency
niceness values.

--
#include <best/regards.h>

Patrick Bellasi

\
 
 \ /
  Last update: 2019-09-05 13:31    [W:0.151 / U:8.824 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site