Date: Wed, 5 Jul 2017 15:20:31 +0200
From: Frederic Weisbecker <fweisbec@gmail.com>
To: Rik van Riel <riel@redhat.com>
Cc: LKML <linux-kernel@vger.kernel.org>, Wanpeng Li <kernellwp@gmail.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Luiz Capitulino <lcapitulino@redhat.com>,
        Ingo Molnar <mingo@kernel.org>
Subject: Re: [PATCH 5/5] sched: Accumulate vtime on top of nsec clocksource
Message-ID: <20170705132030.GD29276@lerouge>
References: <1498756511-11714-1-git-send-email-fweisbec@gmail.com>
 <1498756511-11714-6-git-send-email-fweisbec@gmail.com>
 <1498778847.6130.8.camel@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <1498778847.6130.8.camel@redhat.com>
User-Agent: Mutt/1.5.24 (2015-08-30)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1822
Lines: 44

On Thu, Jun 29, 2017 at 07:27:27PM -0400, Rik van Riel wrote:
> On Thu, 2017-06-29 at 19:15 +0200, Frederic Weisbecker wrote:
> > From: Wanpeng Li <kernellwp@gmail.com>
> > 
> > Currently the cputime source used by vtime is jiffies. When we cross
> > a context boundary and jiffies have changed since the last snapshot,
> > the
> > pending cputime is accounted to the switching out context.
> > 
> > This system works ok if the ticks are not aligned across CPUs. If
> > they
> > instead are aligned (ie: all fire at the same time) and the CPUs run
> > in
> > userspace, the jiffies change is only observed on tick exit and
> > therefore
> > the user cputime is accounted as system cputime. This is because the
> > CPU that maintains timekeeping fires its tick at the same time as the
> > others. It updates jiffies in the middle of the tick and the other
> > CPUs
> > see that update on IRQ exit:
> > 
> > ????CPU 0 (timekeeper)??????????????????CPU 1
> > ????-------------------??????????????-------------
> > ??????????????????????jiffies = N
> > ????...??????????????????????????????run in userspace for a jiffy
> > ????tick entry???????????????????????tick entry (sees jiffies = N)
> > ????set jiffies = N + 1
> > ????tick exit????????????????????????tick exit (sees jiffies = N + 1)
> > ????????????????????????????????????????????????account 1 jiffy as
> > stime
> > 
> > Fix this with using a nanosec clock source instead of jiffies. The
> > cputime is then accumulated and flushed everytime the pending delta
> > reaches a jiffy in order to mitigate the accounting overhead.
> 
> Glad to hear this could be done without dramatically
> increasing the accounting overhead!

Lets hope so, I actually haven't yet measured if there is a
performance delta :-s

If any I don't expect a big one.

Thanks for your reviews!