From: Patrick Bellasi <patrick.bellasi@arm.com>
To: Valentin Schneider <valentin.schneider@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
Subhra Mazumdar <subhra.mazumdar@oracle.com>,
linux-kernel@vger.kernel.org, mingo@redhat.com,
tglx@linutronix.de, steven.sistare@oracle.com,
dhaval.giani@oracle.com, daniel.lezcano@linaro.org,
vincent.guittot@linaro.org, viresh.kumar@linaro.org,
tim.c.chen@linux.intel.com, mgorman@techsingularity.net,
parth@linux.ibm.com
Subject: Re: [RFC PATCH 1/9] sched,cgroup: Add interface for latency-nice
Date: Thu, 05 Sep 2019 14:07:54 +0100 [thread overview]
Message-ID: <87d0ge3n85.fsf@arm.com> (raw)
In-Reply-To: <df69627e-8aa0-e2cb-044e-fb392f34efa5@arm.com>
On Thu, Sep 05, 2019 at 12:46:37 +0100, Valentin Schneider wrote...
> On 05/09/2019 12:18, Patrick Bellasi wrote:
>>> 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.
>>
>> Right, we have this dualism to deal with and current mainline behaviour
>> is somehow in the middle.
>>
>> BTW, the FB requirement is the same we have in Android.
>> We want some CFS tasks to have very small latency and a low chance
>> to be preempted by the wake-up of less-important "background" tasks.
>>
>> I'm not totally against the usage of a signed range, but I'm thinking
>> that since we are introducing a new (non POSIX) concept we can get the
>> chance to make it more human friendly.
>>
>> Give the two extremes above, would not be much simpler and intuitive to
>> have 0 implementing the FB/Android (no latency) case and 1024 the
>> (max latency) Oracle case?
>>
>
> For something like latency-<whatever>, I don't see the point of having
> such a wide range. The nice range is probably more than enough - and before
> even bothering about the range, we should probably agree on what the range
> should represent.
>
> If it's niceness, I read it as: positive latency-nice value means we're
> nice to latency, means we reduce it. So the further up you go, the more you
> restrict your wakeup scan. I think it's quite easy to map that into the
> code: current behaviour at 0, with a decreasing scan mask size as we go
> towards +19. I don't think anyone needs 512 steps to tune this.
>
> I don't know what logic we'd follow for negative values though. Maybe
> latency-nice -20 means always going through the slowpath, but what of the
> intermediate values?
Yep, I think so fare we are all converging towards the idea to use the
a signed range. Regarding the range itself, yes: 1024 looks very
oversized, but +-20 is still something which leave room for a bit of
flexibility and it also better matches the idea that we don't want to
"enumerate behaviours" but just expose a knob. To map certain "bias" we
could benefit from a slightly larger range.
> AFAICT this RFC only looks at wakeups, but I guess latency-nice can be
For the wakeup path there is also the TurboSched proposal by Parth:
Message-ID: <20190725070857.6639-1-parth@linux.ibm.com>
https://lore.kernel.org/lkml/20190725070857.6639-1-parth@linux.ibm.com/
we should keep in mind.
> applied elsewhere (e.g. load-balance, something like task_hot() and its
> use of sysctl_sched_migration_cost).
For LB can you come up with some better description of what usages you
see could benefit from a "per task" or "per task-group" latency niceness?
Best,
Patrick
--
#include <best/regards.h>
Patrick Bellasi
next prev parent reply other threads:[~2019-09-05 13:08 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-30 17:49 [RFC PATCH 0/9] Task latency-nice subhra mazumdar
2019-08-30 17:49 ` [RFC PATCH 1/9] sched,cgroup: Add interface for latency-nice subhra mazumdar
2019-09-04 17:32 ` Tim Chen
2019-09-05 6:15 ` Parth Shah
2019-09-05 10:11 ` Patrick Bellasi
2019-09-06 12:22 ` Parth Shah
2019-09-05 8:31 ` Peter Zijlstra
2019-09-05 9:45 ` Patrick Bellasi
2019-09-05 10:46 ` Peter Zijlstra
2019-09-05 11:13 ` Qais Yousef
2019-09-05 11:30 ` Peter Zijlstra
2019-09-05 11:40 ` Patrick Bellasi
2019-09-05 11:48 ` Peter Zijlstra
2019-09-05 13:32 ` Qais Yousef
2019-09-05 11:47 ` Qais Yousef
2020-04-16 0:02 ` Joel Fernandes
2020-04-16 17:23 ` Dietmar Eggemann
2020-04-18 16:01 ` Joel Fernandes
2020-04-20 11:26 ` Parth Shah
2020-04-20 19:14 ` Joel Fernandes
2020-04-20 11:47 ` Qais Yousef
2020-04-20 19:10 ` Joel Fernandes
2019-09-05 11:30 ` Patrick Bellasi
2019-09-05 11:47 ` Peter Zijlstra
2019-09-05 11:18 ` Patrick Bellasi
2019-09-05 11:40 ` Peter Zijlstra
2019-09-05 11:46 ` Patrick Bellasi
2019-09-05 11:46 ` Valentin Schneider
2019-09-05 13:07 ` Patrick Bellasi [this message]
2019-09-05 14:48 ` Valentin Schneider
2019-09-06 12:45 ` Parth Shah
2019-09-06 14:13 ` Valentin Schneider
2019-09-06 14:32 ` Vincent Guittot
2019-09-06 17:10 ` Parth Shah
2019-09-06 22:50 ` Valentin Schneider
2019-09-06 12:31 ` Parth Shah
2019-09-05 10:05 ` Patrick Bellasi
2019-09-05 10:48 ` Peter Zijlstra
2019-08-30 17:49 ` [RFC PATCH 2/9] sched: add search limit as per latency-nice subhra mazumdar
2019-09-05 6:22 ` Parth Shah
2019-08-30 17:49 ` [RFC PATCH 3/9] sched: add sched feature to disable idle core search subhra mazumdar
2019-09-05 10:17 ` Patrick Bellasi
2019-09-05 22:02 ` Subhra Mazumdar
2019-08-30 17:49 ` [RFC PATCH 4/9] sched: SIS_CORE " subhra mazumdar
2019-09-05 10:19 ` Patrick Bellasi
2019-08-30 17:49 ` [RFC PATCH 5/9] sched: Define macro for number of CPUs in core subhra mazumdar
2019-08-30 17:49 ` [RFC PATCH 6/9] x86/smpboot: Optimize cpumask_weight_sibling macro for x86 subhra mazumdar
2019-08-30 17:49 ` [RFC PATCH 7/9] sched: search SMT before LLC domain subhra mazumdar
2019-09-05 9:31 ` Peter Zijlstra
2019-09-05 20:40 ` Subhra Mazumdar
2019-08-30 17:49 ` [RFC PATCH 8/9] sched: introduce per-cpu var next_cpu to track search limit subhra mazumdar
2019-08-30 17:49 ` [RFC PATCH 9/9] sched: rotate the cpu search window for better spread subhra mazumdar
2019-09-05 6:37 ` Parth Shah
2019-09-05 5:55 ` [RFC PATCH 0/9] Task latency-nice Parth Shah
2019-09-05 10:31 ` Patrick Bellasi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87d0ge3n85.fsf@arm.com \
--to=patrick.bellasi@arm.com \
--cc=daniel.lezcano@linaro.org \
--cc=dhaval.giani@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgorman@techsingularity.net \
--cc=mingo@redhat.com \
--cc=parth@linux.ibm.com \
--cc=peterz@infradead.org \
--cc=steven.sistare@oracle.com \
--cc=subhra.mazumdar@oracle.com \
--cc=tglx@linutronix.de \
--cc=tim.c.chen@linux.intel.com \
--cc=valentin.schneider@arm.com \
--cc=vincent.guittot@linaro.org \
--cc=viresh.kumar@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).