From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-fsdevel-owner@vger.kernel.org>
Received: from mx1.redhat.com ([209.132.183.28]:34076 "EHLO mx1.redhat.com"
        rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
        id S1726581AbfAIS5i (ORCPT <rfc822;linux-fsdevel@vger.kernel.org>);
        Wed, 9 Jan 2019 13:57:38 -0500
Subject: Re: [PATCH v2 0/4] /proc/stat: Reduce irqs counting performance
 overhead
To: Matthew Wilcox <willy@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>,
        Alexey Dobriyan <adobriyan@gmail.com>,
        Kees Cook <keescook@chromium.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
        Davidlohr Bueso <dave@stgolabs.net>,
        Miklos Szeredi <miklos@szeredi.hu>,
        Daniel Colascione <dancol@google.com>,
        Dave Chinner <david@fromorbit.com>,
        Randy Dunlap <rdunlap@infradead.org>
References: <1547054628-12703-1-git-send-email-longman@redhat.com>
 <20190109174403.GN6310@bombadil.infradead.org>
 <f6302369-9eda-ef58-4f18-3dc9ffbc31f1@redhat.com>
 <20190109182417.GO6310@bombadil.infradead.org>
 <f0ae7bd8-5c5b-aefd-aff8-2719fc0edbaa@redhat.com>
 <20190109185226.GP6310@bombadil.infradead.org>
From: Waiman Long <longman@redhat.com>
Message-ID: <116b9b0e-8352-3fed-2dc8-4436915ace57@redhat.com>
Date: Wed, 9 Jan 2019 13:57:36 -0500
MIME-Version: 1.0
In-Reply-To: <20190109185226.GP6310@bombadil.infradead.org>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Content-Language: en-US
Sender: linux-fsdevel-owner@vger.kernel.org
List-ID: <linux-fsdevel.vger.kernel.org>

On 01/09/2019 01:52 PM, Matthew Wilcox wrote:
> On Wed, Jan 09, 2019 at 01:37:23PM -0500, Waiman Long wrote:
>> On 01/09/2019 01:24 PM, Matthew Wilcox wrote:
>>> Did you even try just using the general purpose infrastructure that's
>>> in place?  If that shows a performance problem _then_ it's time to make
>>> this special snowflake just a little more special.  Not before.
>> I have looked into the percpu counter code. There are two aspects that I
>> don't like to introduce to the interrupt handler's code path for
>> updating the counts.
>>
>> 1) There is a raw spinlock in the percpu_counter structure that may need
>> to be acquired in the update path. This can be a performance drag
>> especially if lockdep is enabled.
>>
>> 2) The percpu_counter structure is 40 bytes in size on 64-bit systems
>> compared with just 8 bytes for the percpu count pointer and an
>> additional 4 bytes that I introduced in patch 2. With thousands of irq
>> descriptors, it can consume quite a lot more memory. Memory consumption
>> was a point that you brought up in one of your previous mails.
> Then _argue that_.  Don't just go off and do something random without
> explaining to the rest of us why we're wrong.

Sorry about that. I should have included that in the cover-letter.

Cheers,
Longman