Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Thu, 5 Sep 2019 14:32:01 +0100
From:   Qais Yousef <qais.yousef@arm.com>
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     Patrick Bellasi <patrick.bellasi@arm.com>,
        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
Message-ID: <20190905133200.4w4cupgxgeym3l2k@e107158-lin.cambridge.arm.com>
References: <20190830174944.21741-1-subhra.mazumdar@oracle.com>
 <20190830174944.21741-2-subhra.mazumdar@oracle.com>
 <20190905083127.GA2332@hirez.programming.kicks-ass.net>
 <87r24v2i14.fsf@arm.com>
 <20190905104616.GD2332@hirez.programming.kicks-ass.net>
 <20190905111346.2w6kuqrdvaqvgilu@e107158-lin.cambridge.arm.com>
 <20190905113002.GK2349@hirez.programming.kicks-ass.net>
 <87ftlb2cq6.fsf@arm.com>
 <20190905114802.GN2349@hirez.programming.kicks-ass.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <20190905114802.GN2349@hirez.programming.kicks-ass.net>
User-Agent: NeoMutt/20171215
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 09/05/19 13:48, Peter Zijlstra wrote:
> On Thu, Sep 05, 2019 at 12:40:01PM +0100, Patrick Bellasi wrote:
> > Right, although I think behaviours could still be exported but via a
> > different and configurable interface, using thresholds.
> 
> I would try _really_ hard to avoid pinning down behaviour. The more you
> do that, the less you can change.

While I agree with that but I find there's a contradiction between not
'pinning down behavior' and 'easy and clear way to bias latency sensitive from
end user's perspective'.

Maybe we should protect this with a kconfig + experimental tag until trial
and error show the best way forward?

--
Qais Yousef