Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Sender: Tejun Heo <htejun@gmail.com>
Date:   Sat, 26 Jun 2021 06:06:32 -0400
From:   Tejun Heo <tj@kernel.org>
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     Josh Don <joshdon@google.com>, Ingo Molnar <mingo@redhat.com>,
        Juri Lelli <juri.lelli@redhat.com>,
        Vincent Guittot <vincent.guittot@linaro.org>,
        Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
        Daniel Bristot de Oliveira <bristot@redhat.com>,
        Paul Turner <pjt@google.com>,
        David Rientjes <rientjes@google.com>,
        Oleg Rombakh <olegrom@google.com>,
        Viresh Kumar <viresh.kumar@linaro.org>,
        Steve Sistare <steven.sistare@oracle.com>,
        linux-kernel@vger.kernel.org, Rik van Riel <riel@surriel.com>
Subject: Re: [PATCH] sched: cgroup SCHED_IDLE support
Message-ID: <YNb8KI5zww8Pweat@mtj.duckdns.org>
References: <20210608231132.32012-1-joshdon@google.com>
 <YMobzbLecaFYuLtq@slm.duckdns.org>
 <YNWPBDH/mQdojEu9@hirez.programming.kicks-ass.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <YNWPBDH/mQdojEu9@hirez.programming.kicks-ass.net>
Precedence: bulk

Hello, Peter.

On Fri, Jun 25, 2021 at 10:08:36AM +0200, Peter Zijlstra wrote:
> It's a direct concequence of the hierarchical requirement. The approach
> is the only valid one. The other relative controllers that don't do
> this, are simply broken.
> 
> Absolute controllers have it easier, they can be trivially flattened.

That's too strong a claim. e.g. iocost controller, while in a different
domain, is a weight controller which takes different trade-offs to achieve
hierarchical weight based distribution at negligible nesting overhead. There
usually are more than one way to skin a cat.

> > There are several practical challenges with the current implementation
> > caused by the full nesting - e.g. nesting levels are expensive for context
> > switch heavy applicaitons often going over >1% per level,
> 
> Yeah, and there's numerical problems you run into as well due to
> limitied precision.

Another issue is per-queue level heuristics like boosting after idle
nesting not in quite optimal ways.

> Just don't do deep hierarchies.
>
> AFAICT it's a simple matter of conflicting requirements, on the one hand
> the hierarchical thing is required, on the other hand people seem to
> think all this crap is 'free' and create super deep hierarchies and then
> complain shit don't work right.

The problem is that the overhead is significant enough even at pretty
shallow levels. Even at just two/three levels, the cost is already
significant enough for some large-scale applications.

Thanks.

-- 
tejun