linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Tejun Heo <tj@kernel.org>
To: Peter Zijlstra <peterz@infradead.org>
Cc: joel@joelfernandes.org, chris.hyser@oracle.com,
	joshdon@google.com, mingo@kernel.org, vincent.guittot@linaro.org,
	valentin.schneider@arm.com, mgorman@suse.de,
	linux-kernel@vger.kernel.org, tglx@linutronix.de,
	"Michal Koutný" <mkoutny@suse.com>,
	"Christian Brauner" <christian.brauner@ubuntu.com>,
	"Zefan Li" <lizefan.x@bytedance.com>
Subject: Re: [PATCH 0/9] sched: Core scheduling interfaces
Date: Tue, 6 Apr 2021 12:08:50 -0400	[thread overview]
Message-ID: <YGyHknFJhHO99e5a@slm.duckdns.org> (raw)
In-Reply-To: <YGx+9EVS5kDrLOD0@hirez.programming.kicks-ass.net>

Hello,

On Tue, Apr 06, 2021 at 05:32:04PM +0200, Peter Zijlstra wrote:
> > I find it difficult to like the proposed interface from the name (the term
> > "core" is really confusing given how the word tends to be used internally)
> > to the semantics (it isn't like anything else) and even the functionality
> > (we're gonna have fixed processors at some point, right?).
> 
> Core is the topological name for the thing that hosts the SMT threads.
> Can't really help that.

I find the name pretty unfortunate given how overloaded the term is
generally and also in kernel but oh well...

> > Here are some preliminary thoughts:
> > 
> > * Are both prctl and cgroup based interfaces really necessary? I could be
> >   being naive but given that we're (hopefully) working around hardware
> >   deficiencies which will go away in time, I think there's a strong case for
> >   minimizing at least the interface to the bare minimum.
> 
> I'm not one for cgroups much, so I'll let others argue that case, except
> that per systemd and all the other new fangled shit, people seem to use
> cgroups a lot to group tasks. So it makes sense to also expose this
> through cgroups in some form.

All the new fangled things follow a certain usage pattern which makes
aligning parts of process tree with cgroup layout trivial, so when
restrictions can be applied along the process tree like this and there isn't
a particular need for dynamic hierarchical control, there isn't much need
for direct cgroup interface.

> That said; I've had requests from lots of non security folks about this
> feature to help mitigate the SMT interference.
> 
> Consider for example Real-Time. If you have an active SMT sibling, the
> CPU performance is much less than it would be when the SMT sibling is
> idle. Therefore, for the benefit of determinism, it would be very nice
> if RT tasks could force-idle their SMT siblings, and voila, this
> interface allows exactly that.
> 
> The same is true for other workloads that care about interference.

I see.

> >   Given how cgroups are set up (membership operations happening only for
> >   seeding, especially with the new clone interface), it isn't too difficult
> >   to synchronize process tree and cgroup hierarchy where it matters - ie.
> >   given the right per-process level interface, restricting configuration for
> >   a cgroup sub-hierarchy may not need any cgroup involvement at all. This
> >   also nicely gets rid of the interaction between prctl and cgroup bits.
> 
> I've no idea what you mean :/ The way I use cgroups (when I have to, for
> testing) is to echo the pid into /cgroup/foo/tasks. No clone or anything
> involved.

The usage pattern is creating a new cgroup, seeding it with a process
(either writing to tasks or using CLONE_INTO_CGROUP) and let it continue
only on that sub-hierarchy, so cgroup hierarchy usually partially overlays
process trees.

> None of my test machines come up with cgroupfs mounted, and any and all
> cgroup setup is under my control.
>
> > * If we *have* to have cgroup interface, I wonder whether this would fit a
> >   lot better as a part of cpuset. If you squint just right, this can be
> >   viewed as some dynamic form of cpuset. Implementation-wise, it probably
> >   won't integrate with the rest but I think the feature will be less jarring
> >   as a part of cpuset, which already is a bit of kitchensink anyway.
> 
> Not sure I agree, we do not change the affinity of things, we only
> control who's allowed to run concurrently on SMT siblings. There could
> be a cpuset partition split between the siblings and it would still work
> fine.

I see. Yeah, if we really need it, I'm not sure it fits in cgroup interface
proper. As I wrote elsewhere, these things are usually implemented on the
originating subsystem interface with cgroup ID as a parameter.

Thanks.

-- 
tejun

  reply	other threads:[~2021-04-06 16:08 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-01 13:10 Peter Zijlstra
2021-04-01 13:10 ` [PATCH 1/9] sched: Allow sched_core_put() from atomic context Peter Zijlstra
2021-04-01 13:10 ` [PATCH 2/9] sched: Implement core-sched assertions Peter Zijlstra
2021-04-01 13:10 ` [PATCH 3/9] sched: Trivial core scheduling cookie management Peter Zijlstra
2021-04-01 20:04   ` Josh Don
2021-04-02  7:13     ` Peter Zijlstra
2021-04-01 13:10 ` [PATCH 4/9] sched: Default core-sched policy Peter Zijlstra
2021-04-21 13:33   ` Peter Zijlstra
2021-04-21 14:31     ` Chris Hyser
2021-04-01 13:10 ` [PATCH 5/9] sched: prctl() core-scheduling interface Peter Zijlstra
2021-04-07 17:00   ` Peter Zijlstra
2021-04-18  3:52     ` Joel Fernandes
2021-04-01 13:10 ` [PATCH 6/9] kselftest: Add test for core sched prctl interface Peter Zijlstra
2021-04-01 13:10 ` [PATCH 7/9] sched: Cgroup core-scheduling interface Peter Zijlstra
2021-04-02  0:34   ` Josh Don
2021-04-01 13:10 ` [PATCH 8/9] rbtree: Remove const from the rb_find_add() comparator Peter Zijlstra
2021-04-01 13:10 ` [PATCH 9/9] sched: prctl() and cgroup interaction Peter Zijlstra
2021-04-03  1:30   ` Josh Don
2021-04-06 15:12     ` Peter Zijlstra
2021-04-04 23:39 ` [PATCH 0/9] sched: Core scheduling interfaces Tejun Heo
2021-04-05 18:46   ` Joel Fernandes
2021-04-06 14:16     ` Tejun Heo
2021-04-18  1:35       ` Joel Fernandes
2021-04-19  9:00         ` Peter Zijlstra
2021-04-21 13:35           ` Peter Zijlstra
2021-04-21 14:45             ` Chris Hyser
2021-04-06 15:32   ` Peter Zijlstra
2021-04-06 16:08     ` Tejun Heo [this message]
2021-04-07 18:39       ` Peter Zijlstra
2021-04-07 16:50   ` Michal Koutný
2021-04-07 18:34     ` Peter Zijlstra
2021-04-08 13:25       ` Michal Koutný
2021-04-08 15:02         ` Peter Zijlstra
2021-04-09  0:16           ` Josh Don
2021-04-19 11:30       ` Tejun Heo
2021-04-20  1:17         ` Josh Don

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=YGyHknFJhHO99e5a@slm.duckdns.org \
    --to=tj@kernel.org \
    --cc=chris.hyser@oracle.com \
    --cc=christian.brauner@ubuntu.com \
    --cc=joel@joelfernandes.org \
    --cc=joshdon@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lizefan.x@bytedance.com \
    --cc=mgorman@suse.de \
    --cc=mingo@kernel.org \
    --cc=mkoutny@suse.com \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --cc=valentin.schneider@arm.com \
    --cc=vincent.guittot@linaro.org \
    --subject='Re: [PATCH 0/9] sched: Core scheduling interfaces' \
    /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

This is a public inbox, see mirroring instructions
on how to clone and mirror all data and code used for this inbox