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: <1565329531-12327-1-git-send-email-wanpengli@tencent.com>
 <fad8ceed-8b98-8fc4-5b6a-63bbca4059a8@redhat.com> <CANRm+CwMMUEyZXmiUu5Y8GA=BEUYGLw31CRyZTc2uA+ct4Bamg@mail.gmail.com>
 <ba07fb02-9b55-15e4-d240-24da59e09369@redhat.com>
In-Reply-To: <ba07fb02-9b55-15e4-d240-24da59e09369@redhat.com>
From:   Wanpeng Li <kernellwp@gmail.com>
Date:   Thu, 15 Aug 2019 12:04:48 +0800
Message-ID: <CANRm+CzGNyST4=BtE-eKvjB-PUVVoM-gUC2Np8NH7tm0Gp2_nQ@mail.gmail.com>
Subject: Re: [PATCH] KVM: LAPIC: Periodically revaluate appropriate lapic_timer_advance_ns
To:     Paolo Bonzini <pbonzini@redhat.com>
Cc:     LKML <linux-kernel@vger.kernel.org>, kvm <kvm@vger.kernel.org>,
        =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= <rkrcmar@redhat.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, 14 Aug 2019 at 20:50, Paolo Bonzini <pbonzini@redhat.com> wrote:
>
> On 12/08/19 11:06, Wanpeng Li wrote:
> > On Fri, 9 Aug 2019 at 18:24, Paolo Bonzini <pbonzini@redhat.com> wrote:
> >>
> >> On 09/08/19 07:45, Wanpeng Li wrote:
> >>> From: Wanpeng Li <wanpengli@tencent.com>
> >>>
> >>> Even if for realtime CPUs, cache line bounces, frequency scaling, presence
> >>> of higher-priority RT tasks, etc can cause different response. These
> >>> interferences should be considered and periodically revaluate whether
> >>> or not the lapic_timer_advance_ns value is the best, do nothing if it is,
> >>> otherwise recaluate again.
> >>
> >> How much fluctuation do you observe between different runs?
> >
> > Sometimes can ~1000 cycles after converting to guest tsc freq.
>
> Hmm, I wonder if we need some kind of continuous smoothing.  Something like

Actually this can fluctuate drastically instead of continuous
smoothing during testing (running linux guest instead of
kvm-unit-tests).

>
>         if (abs(advance_expire_delta) < LAPIC_TIMER_ADVANCE_ADJUST_DONE) {
>                 /* no update for random fluctuations */
>                 return;
>         }
>
>         if (unlikely(timer_advance_ns > 5000))
>                 timer_advance_ns = LAPIC_TIMER_ADVANCE_ADJUST_INIT;
>         apic->lapic_timer.timer_advance_ns = timer_advance_ns;
>
> and removing all the timer_advance_adjust_done stuff.  What do you think?

I just sent out v2, periodically revaluate and get a minimal
conservative value from these revaluate points. Please have a look. :)

Regards,
Wanpeng Li