From: Peter Zijlstra <peterz@infradead.org>
To: Patrick Bellasi <patrick.bellasi@arm.com>
Cc: linux-kernel@vger.kernel.org, linux-pm@vger.kernel.org,
linux-api@vger.kernel.org, Ingo Molnar <mingo@redhat.com>,
Tejun Heo <tj@kernel.org>,
"Rafael J . Wysocki" <rafael.j.wysocki@intel.com>,
Vincent Guittot <vincent.guittot@linaro.org>,
Viresh Kumar <viresh.kumar@linaro.org>,
Paul Turner <pjt@google.com>,
Quentin Perret <quentin.perret@arm.com>,
Dietmar Eggemann <dietmar.eggemann@arm.com>,
Morten Rasmussen <morten.rasmussen@arm.com>,
Juri Lelli <juri.lelli@redhat.com>, Todd Kjos <tkjos@google.com>,
Joel Fernandes <joelaf@google.com>,
Steve Muckle <smuckle@google.com>,
Suren Baghdasaryan <surenb@google.com>
Subject: Re: [PATCH v6 09/16] sched/cpufreq: uclamp: Add utilization clamping for RT tasks
Date: Wed, 23 Jan 2019 21:11:46 +0100 [thread overview]
Message-ID: <20190123201146.GH17749@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <20190123144011.iid3avb63r5v4r2c@e110439-lin>
On Wed, Jan 23, 2019 at 02:40:11PM +0000, Patrick Bellasi wrote:
> On 23-Jan 11:49, Peter Zijlstra wrote:
> > On Tue, Jan 15, 2019 at 10:15:06AM +0000, Patrick Bellasi wrote:
> > > @@ -858,16 +859,23 @@ static inline void
> > > uclamp_effective_get(struct task_struct *p, unsigned int clamp_id,
> > > unsigned int *clamp_value, unsigned int *bucket_id)
> > > {
> > > + struct uclamp_se *default_clamp;
> > > +
> > > /* Task specific clamp value */
> > > *clamp_value = p->uclamp[clamp_id].value;
> > > *bucket_id = p->uclamp[clamp_id].bucket_id;
> > >
> > > + /* RT tasks have different default values */
> > > + default_clamp = task_has_rt_policy(p)
> > > + ? uclamp_default_perf
> > > + : uclamp_default;
> > > +
> > > /* System default restriction */
> > > - if (unlikely(*clamp_value < uclamp_default[UCLAMP_MIN].value ||
> > > - *clamp_value > uclamp_default[UCLAMP_MAX].value)) {
> > > + if (unlikely(*clamp_value < default_clamp[UCLAMP_MIN].value ||
> > > + *clamp_value > default_clamp[UCLAMP_MAX].value)) {
> > > /* Keep it simple: unconditionally enforce system defaults */
> > > - *clamp_value = uclamp_default[clamp_id].value;
> > > - *bucket_id = uclamp_default[clamp_id].bucket_id;
> > > + *clamp_value = default_clamp[clamp_id].value;
> > > + *bucket_id = default_clamp[clamp_id].bucket_id;
> > > }
> > > }
> >
> > So I still don't much like the whole effective thing;
>
> :/
>
> I find back-annotation useful in many cases since we have different
> sources for possible clamp values:
>
> 1. task specific
> 2. cgroup defined
> 3. system defaults
> 4. system power default
(I'm not sure I've seen 4 happen yet, what's that?)
Anyway, once you get range composition defined; that should be something
like:
R_p \Compose_g R_g
Where R_p is the range of task-p, and R_g is the range of the g'th
cgroup of p (where you can make an identity between the root cgroup and
the system default).
Now; as per the other email; I think the straight forward composition:
struct range compose(struct range a, struct range b)
{
return (range){.min = clamp(a.min, b.min, b.max),
.max = clamp(a.max, b.min, b.max), };
}
(note that this is non-commutative, so we have to pay attention to
get the order 'right')
Works in this case; unlike the cpu/rq conposition where we resort to a
pure max function for non-interference.
> I don't think we can avoid to somehow back annotate on which bucket a
> task has been refcounted... it makes dequeue so much easier, it helps
> in ensuring that the refcouning is consistent and enable lazy updates.
So I'll have to go over the code again, but I'm wondering why you're
changing uclamp_se::bucket_id on a runnable task.
Ideally you keep bucket_id invariant between enqueue and dequeue; then
dequeue knows where we put it.
Now I suppose actually determining bucket_id is 'expensive' (it
certainly is with the whole mapping scheme, but even that integer
division is not nice), so we'd like to precompute the bucket_id.
This then leads to the problem of how to change uclamp_se::value while
runnable (since per the other thread you don't want to always update all
runnable tasks).
So far so right?
I'm thikning that if we haz a single bit, say:
struct uclamp_se {
...
unsigned int changed : 1;
};
We can update uclamp_se::value and set uclamp_se::changed, and then the
next enqueue will (unlikely) test-and-clear changed and recompute the
bucket_id.
Would that not be simpler?
next prev parent reply other threads:[~2019-01-23 20:11 UTC|newest]
Thread overview: 89+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-15 10:14 [PATCH v6 00/16] Add utilization clamping support Patrick Bellasi
2019-01-15 10:14 ` [PATCH v6 01/16] sched/core: Allow sched_setattr() to use the current policy Patrick Bellasi
2019-01-25 13:56 ` Alessio Balsini
2019-01-15 10:14 ` [PATCH v6 02/16] sched/core: uclamp: Extend sched_setattr() to support utilization clamping Patrick Bellasi
2019-01-15 10:15 ` [PATCH v6 03/16] sched/core: uclamp: Map TASK's clamp values into CPU's clamp buckets Patrick Bellasi
2019-01-21 10:15 ` Peter Zijlstra
2019-01-21 12:27 ` Patrick Bellasi
2019-01-21 12:51 ` Peter Zijlstra
2019-01-21 15:05 ` Peter Zijlstra
2019-01-21 15:34 ` Patrick Bellasi
2019-01-15 10:15 ` [PATCH v6 04/16] sched/core: uclamp: Add CPU's clamp buckets refcounting Patrick Bellasi
2019-01-21 14:59 ` Peter Zijlstra
2019-01-21 15:23 ` Patrick Bellasi
2019-01-21 16:12 ` Peter Zijlstra
2019-01-21 16:33 ` Patrick Bellasi
2019-01-22 9:45 ` Peter Zijlstra
2019-01-22 10:31 ` Patrick Bellasi
2019-01-21 15:17 ` Peter Zijlstra
2019-01-21 15:54 ` Patrick Bellasi
2019-01-22 10:03 ` Peter Zijlstra
2019-01-22 10:53 ` Patrick Bellasi
2019-01-15 10:15 ` [PATCH v6 05/16] sched/core: uclamp: Update CPU's refcount on clamp changes Patrick Bellasi
2019-01-21 15:33 ` Peter Zijlstra
2019-01-21 15:44 ` Patrick Bellasi
2019-01-22 9:37 ` Peter Zijlstra
2019-01-22 10:43 ` Patrick Bellasi
2019-01-22 13:28 ` Peter Zijlstra
2019-01-22 14:01 ` Patrick Bellasi
2019-01-22 14:57 ` Peter Zijlstra
2019-01-22 15:33 ` Patrick Bellasi
2019-01-23 9:16 ` Peter Zijlstra
2019-01-23 14:14 ` Patrick Bellasi
2019-01-23 18:59 ` Peter Zijlstra
2019-01-24 11:21 ` Patrick Bellasi
2019-01-24 12:38 ` Peter Zijlstra
2019-01-15 10:15 ` [PATCH v6 06/16] sched/core: uclamp: Enforce last task UCLAMP_MAX Patrick Bellasi
2019-01-15 10:15 ` [PATCH v6 07/16] sched/core: uclamp: Add system default clamps Patrick Bellasi
2019-01-22 13:56 ` Peter Zijlstra
2019-01-22 14:43 ` Patrick Bellasi
2019-01-22 15:13 ` Peter Zijlstra
2019-01-22 15:41 ` Patrick Bellasi
2019-01-23 9:22 ` Peter Zijlstra
2019-01-23 14:19 ` Patrick Bellasi
2019-01-23 19:10 ` Peter Zijlstra
2019-01-15 10:15 ` [PATCH v6 08/16] sched/cpufreq: uclamp: Add utilization clamping for FAIR tasks Patrick Bellasi
2019-01-22 10:37 ` Rafael J. Wysocki
2019-01-22 11:02 ` Patrick Bellasi
2019-01-22 11:04 ` Rafael J. Wysocki
2019-01-22 11:27 ` Patrick Bellasi
2019-01-22 15:21 ` Peter Zijlstra
2019-01-22 15:45 ` Patrick Bellasi
2019-01-22 17:13 ` Peter Zijlstra
2019-01-22 18:18 ` Patrick Bellasi
2019-01-23 9:52 ` Peter Zijlstra
2019-01-23 14:24 ` Patrick Bellasi
2019-01-15 10:15 ` [PATCH v6 09/16] sched/cpufreq: uclamp: Add utilization clamping for RT tasks Patrick Bellasi
2019-01-22 12:30 ` Quentin Perret
2019-01-22 12:37 ` Patrick Bellasi
2019-01-23 10:28 ` Peter Zijlstra
2019-01-23 14:33 ` Patrick Bellasi
2019-01-23 10:49 ` Peter Zijlstra
2019-01-23 14:40 ` Patrick Bellasi
2019-01-23 20:11 ` Peter Zijlstra [this message]
2019-01-24 12:30 ` Patrick Bellasi
2019-01-24 12:38 ` Patrick Bellasi
2019-01-24 15:12 ` Peter Zijlstra
2019-01-24 16:00 ` Patrick Bellasi
2019-01-24 15:31 ` Peter Zijlstra
2019-01-24 16:14 ` Patrick Bellasi
2019-01-24 15:33 ` Peter Zijlstra
2019-01-24 15:15 ` Peter Zijlstra
2019-01-24 16:05 ` Patrick Bellasi
2019-01-15 10:15 ` [PATCH v6 10/16] sched/core: Add uclamp_util_with() Patrick Bellasi
2019-01-23 13:33 ` Peter Zijlstra
2019-01-23 14:51 ` Patrick Bellasi
2019-01-23 19:22 ` Peter Zijlstra
2019-01-15 10:15 ` [PATCH v6 11/16] sched/fair: Add uclamp support to energy_compute() Patrick Bellasi
2019-01-22 12:13 ` Quentin Perret
2019-01-22 12:45 ` Patrick Bellasi
2019-01-22 13:29 ` Quentin Perret
2019-01-22 14:26 ` Patrick Bellasi
2019-01-22 14:39 ` Quentin Perret
2019-01-22 15:01 ` Patrick Bellasi
2019-01-22 15:14 ` Quentin Perret
2019-01-15 10:15 ` [PATCH v6 12/16] sched/core: uclamp: Extend CPU's cgroup controller Patrick Bellasi
2019-01-15 10:15 ` [PATCH v6 13/16] sched/core: uclamp: Propagate parent clamps Patrick Bellasi
2019-01-15 10:15 ` [PATCH v6 14/16] sched/core: uclamp: Map TG's clamp values into CPU's clamp buckets Patrick Bellasi
2019-01-15 10:15 ` [PATCH v6 15/16] sched/core: uclamp: Use TG's clamps to restrict TASK's clamps Patrick Bellasi
2019-01-15 10:15 ` [PATCH v6 16/16] sched/core: uclamp: Update CPU's refcount on TG's clamp changes 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=20190123201146.GH17749@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=dietmar.eggemann@arm.com \
--cc=joelaf@google.com \
--cc=juri.lelli@redhat.com \
--cc=linux-api@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=morten.rasmussen@arm.com \
--cc=patrick.bellasi@arm.com \
--cc=pjt@google.com \
--cc=quentin.perret@arm.com \
--cc=rafael.j.wysocki@intel.com \
--cc=smuckle@google.com \
--cc=surenb@google.com \
--cc=tj@kernel.org \
--cc=tkjos@google.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).