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;
Date:   Thu, 17 Feb 2022 14:45:24 +0000
From:   Aaron Tomlin <atomlin@atomlin.com>
To:     Frederic Weisbecker <frederic@kernel.org>
Cc:     Aaron Tomlin <atomlin@redhat.com>, Phil Auld <pauld@redhat.com>,
        tglx@linutronix.de, mingo@kernel.org, mingo@redhat.com,
        linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [RFC PATCH] tick/sched: Ensure quiet_vmstat() is called when the
 idle tick was stopped too
Message-ID: <20220217144524.el7lwi45hwsf3ef2@ava.usersys.com>
References: <20220203214339.1889971-1-atomlin@redhat.com>
 <YfxVpEO+UJTC+a9e@lorien.usersys.redhat.com>
 <20220216143412.dwxjlkq4w2zeweld@ava.usersys.com>
 <20220217125731.GA744754@lothringen>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <20220217125731.GA744754@lothringen>
Precedence: bulk

On Thu 2022-02-17 13:57 +0100, Frederic Weisbecker wrote:
> Hmm, but does it matter? The issue seem to be that we can enter in idle loop without
> flushing vmstat. Or am I missing something else?
Frederic,

Yes, this is indeed the concern. So, the idea I had was to invoke
quiet_vmstat() regardless if the tick was stopped or not. If I understand
correctly, this should resolve the issue. Furthermore, folding of all
outstanding differentials will only occur when required. Thus performance
should be negligible.


Kind regards,

-- 
Aaron Tomlin