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;
MIME-Version: 1.0
References: <1546873978-27797-1-git-send-email-longman@redhat.com> <20190107223214.GZ6311@dastard>
In-Reply-To: <20190107223214.GZ6311@dastard>
From:   Daniel Colascione <dancol@google.com>
Date:   Mon, 7 Jan 2019 17:41:28 -0500
Message-ID: <CAKOZueuHGg=F8se1AQqswZWCEco=ScmudWq5X5s1qS4+LCtgWw@mail.gmail.com>
Subject: Re: [PATCH 0/2] /proc/stat: Reduce irqs counting performance overhead
To:     Dave Chinner <david@fromorbit.com>
Cc:     Waiman Long <longman@redhat.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Alexey Dobriyan <adobriyan@gmail.com>,
        Luis Chamberlain <mcgrof@kernel.org>,
        Kees Cook <keescook@chromium.org>,
        Jonathan Corbet <corbet@lwn.net>,
        linux-kernel <linux-kernel@vger.kernel.org>,
        "open list:DOCUMENTATION" <linux-doc@vger.kernel.org>,
        Linux FS Devel <linux-fsdevel@vger.kernel.org>,
        Davidlohr Bueso <dave@stgolabs.net>,
        Miklos Szeredi <miklos@szeredi.hu>,
        Randy Dunlap <rdunlap@infradead.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Jan 7, 2019 at 5:32 PM Dave Chinner <david@fromorbit.com> wrote:
>
> On Mon, Jan 07, 2019 at 10:12:56AM -0500, Waiman Long wrote:
> > As newer systems have more and more IRQs and CPUs available in their
> > system, the performance of reading /proc/stat frequently is getting
> > worse and worse.
>
> Because the "roll-your-own" per-cpu counter implementaiton has been
> optimised for low possible addition overhead on the premise that
> summing the counters is rare and isn't a performance issue. This
> patchset is a direct indication that this "summing is rare and can
> be slow" premise is now invalid.

Focusing on counter performance is, IMHO, missing the mark. Even if
interrupt count collection were made fast, there's *something* in any
particular /proc file that a particular reader doesn't need and that,
by being uselessly collected, needlessly slows that reader. There
should be a general-purpose way for /proc file readers to tell the
kernel which bits of information interest them on a particular read
syscall sequence or particular open(2) or something. Creating a new
proc file for every useful combination of attributes doesn't scale
either.