Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
From:   Joel Fernandes <joel@joelfernandes.org>
Mime-Version: 1.0 (1.0)
Subject: Re: [RFC PATCH 07/11] sched: Add proxy execution
Date:   Mon, 17 Oct 2022 00:03:09 -0400
Message-Id: <F34EC553-9C1E-4027-A4BB-BAA8D443235F@joelfernandes.org>
References: <20221016164809.6cd6871f@rorschach.local.home>
Cc:     Peter Zijlstra <peterz@infradead.org>,
        Connor O'Brien <connoro@google.com>,
        linux-kernel@vger.kernel.org, kernel-team@android.com,
        John Stultz <jstultz@google.com>,
        Qais Yousef <qais.yousef@arm.com>,
        Ingo Molnar <mingo@redhat.com>,
        Juri Lelli <juri.lelli@redhat.com>,
        Vincent Guittot <vincent.guittot@linaro.org>,
        Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
        Daniel Bristot de Oliveira <bristot@redhat.com>,
        Valentin Schneider <vschneid@redhat.com>,
        Will Deacon <will@kernel.org>,
        Waiman Long <longman@redhat.com>,
        Boqun Feng <boqun.feng@gmail.com>,
        "Paul E . McKenney" <paulmck@kernel.org>, youssefesmat@google.com
In-Reply-To: <20221016164809.6cd6871f@rorschach.local.home>
To:     Steven Rostedt <rostedt@goodmis.org>
Precedence: bulk


> On Oct 16, 2022, at 4:48 PM, Steven Rostedt <rostedt@goodmis.org> wrote:
>=20
> =EF=BB=BFOn Sat, 15 Oct 2022 15:53:19 +0200
> Peter Zijlstra <peterz@infradead.org> wrote:
>=20
>>> =46rom this it is easy to see that the critical secion executes with the=
 =20
>> direct sum of the blockchain as a whole (all of them will have donated
>> their relative time to make the owner elegible again) -- provided the
>> critical section is of course long enough for this to matter (or it's
>> owner's weight small enough etc..).
>=20
> Does this mean that a lower priority task could do a sort of DOS attack
> on a high priority task, if it creates a bunch of threads that
> constantly grabs a shared lock from the higher priority task? That is,
> the higher priority task could possibly lose a lot of its quota due to
> other tasks running on its behalf in the critical section?

Just my opinion: I can see that as a problem, but on the other hand, if a bu=
nch of threads are acquiring a shared lock to block a high priority task, th=
en they are probably causing a DOS-type of issue anyway whether PE is enable=
d or not.  Even if a prio inheritance mechanism boosts them, they=E2=80=99re=
 still making the progress of the blocking high priority tasks slower. Or, d=
id I miss something in this use case?

Thanks.