From mboxrd@z Thu Jan  1 00:00:00 1970
From: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Subject: Re: [PATCH] cpufreq: intel_pstate: Improve IO performance
Date: Thu, 03 Aug 2017 18:47:09 -0700
Message-ID: <1501811229.4920.93.camel@linux.intel.com>
References: <1501645506-29398-1-git-send-email-srinivas.pandruvada@linux.intel.com>
         <CAJZ5v0henjb2Y0rJdc92BJVgGHg1Nb=eWLnd75eVBP5PJVmCog@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8bit
Return-path: <linux-pm-owner@vger.kernel.org>
Received: from mga14.intel.com ([192.55.52.115]:43967 "EHLO mga14.intel.com"
        rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
        id S1751811AbdHDBrK (ORCPT <rfc822;linux-pm@vger.kernel.org>);
        Thu, 3 Aug 2017 21:47:10 -0400
In-Reply-To: <CAJZ5v0henjb2Y0rJdc92BJVgGHg1Nb=eWLnd75eVBP5PJVmCog@mail.gmail.com>
Sender: linux-pm-owner@vger.kernel.org
List-Id: linux-pm@vger.kernel.org
To: "Rafael J. Wysocki" <rafael@kernel.org>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>, Len Brown <lenb@kernel.org>, Linux PM <linux-pm@vger.kernel.org>

On Fri, 2017-08-04 at 02:34 +0200, Rafael J. Wysocki wrote:
> On Wed, Aug 2, 2017 at 5:45 AM, Srinivas Pandruvada
> <srinivas.pandruvada@linux.intel.com> wrote:
> > 
> > In the current implementation the latency from SCHED_CPUFREQ_IOWAIT
> > is
> > set to actual P-state adjustment can be upto 10ms. This can be
> > improved
> > by reacting to SCHED_CPUFREQ_IOWAIT by jumping to max P-state
> > immediately
> > . With this change the IO performance improves significantly.
> > 
> > With a simple "grep -r . linux" (Here linux is kernel source
> > folder) with
> > dropped caches every time on a platform with per core P-states on a
> > Broadwell Xeon workstation, the user and system time improves as
> > much as
> > 30% to 40%.
> > 
> > The same performance difference was not observed on clients, which
> > don't
> > have per core P-state support.
> > 
> > Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel
> > .com>
> > ---
> >  drivers/cpufreq/intel_pstate.c | 10 ++++++++++
> >  1 file changed, 10 insertions(+)
> > 
> > diff --git a/drivers/cpufreq/intel_pstate.c
> > b/drivers/cpufreq/intel_pstate.c
> > index 8c67b77..7762255 100644
> > --- a/drivers/cpufreq/intel_pstate.c
> > +++ b/drivers/cpufreq/intel_pstate.c
> > @@ -1527,6 +1527,15 @@ static void intel_pstate_update_util(struct
> > update_util_data *data, u64 time,
> > 
> >         if (flags & SCHED_CPUFREQ_IOWAIT) {
> >                 cpu->iowait_boost = int_tofp(1);
> > +               /*
> > +                * The last time the busy was 100% so P-state was
> > max anyway
> > +                * so avoid overhead of computation.
> > +                */
> > +               if (fp_toint(cpu->sample.busy_scaled) == 100) {
> > +                       cpu->last_update = time;
> > +                       return;
> > +               }
> > +               goto set_pstate;
> cpu->last_update should also be updated when you jump to set_pstate,
> shouldn't it?
Yes. It should be updated.

Thanks,
Srinivas

> 
> > 
> >         } else if (cpu->iowait_boost) {
> >                 /* Clear iowait_boost if the CPU may have been
> > idle. */
> >                 delta_ns = time - cpu->last_update;
> > @@ -1538,6 +1547,7 @@ static void intel_pstate_update_util(struct
> > update_util_data *data, u64 time,
> >         if ((s64)delta_ns < INTEL_PSTATE_DEFAULT_SAMPLING_INTERVAL)
> >                 return;
> > 
> > +set_pstate:
> >         if (intel_pstate_sample(cpu, time)) {
> >                 int target_pstate;
> > 
> > --
> > 2.7.4
> >