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;
From:   Francisco Jerez <currojerez@riseup.net>
To:     "Pandruvada\, Srinivas" <srinivas.pandruvada@intel.com>,
        "lenb\@kernel.org" <lenb@kernel.org>,
        "mgorman\@techsingularity.net" <mgorman@techsingularity.net>,
        "rjw\@rjwysocki.net" <rjw@rjwysocki.net>,
        "Tamminen\, Eero T" <eero.t.tamminen@intel.com>
Cc:     "linux-kernel\@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "peterz\@infradead.org" <peterz@infradead.org>,
        "ggherdovich\@suse.cz" <ggherdovich@suse.cz>,
        "linux-pm\@vger.kernel.org" <linux-pm@vger.kernel.org>,
        "juri.lelli\@redhat.com" <juri.lelli@redhat.com>,
        "viresh.kumar\@linaro.org" <viresh.kumar@linaro.org>,
        "joonas.lahtinen\@linux.intel.com" <joonas.lahtinen@linux.intel.com>,
        "tvrtko.ursulin\@linux.intel.com" <tvrtko.ursulin@linux.intel.com>,
        "chris\@chris-wilson.co.uk" <chris@chris-wilson.co.uk>
Subject: Re: [PATCH 4/4] cpufreq: intel_pstate: enable boost for Skylake Xeon
In-Reply-To: <9828ba535fcdce8458593013fd1c67385a8fefb9.camel@intel.com>
References: <20180605214242.62156-1-srinivas.pandruvada@linux.intel.com> <20180605214242.62156-5-srinivas.pandruvada@linux.intel.com> <87bmarhqk4.fsf@riseup.net> <3afad4b13370fe0ba6ed5020f86ce1b8ff46c603.camel@linux.intel.com> <9828ba535fcdce8458593013fd1c67385a8fefb9.camel@intel.com>
Date:   Sat, 28 Jul 2018 13:23:50 -0700
Message-ID: <87muubf6sp.fsf@riseup.net>
MIME-Version: 1.0
Content-Type: multipart/signed; boundary="==-=-=";
        micalg=pgp-sha256; protocol="application/pgp-signature"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

--==-=-=
Content-Type: multipart/mixed; boundary="=-=-="

--=-=-=
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

"Pandruvada, Srinivas" <srinivas.pandruvada@intel.com> writes:

> On Sat, 2018-07-28 at 07:14 -0700, Srinivas Pandruvada wrote:
>> On Fri, 2018-07-27 at 22:34 -0700, Francisco Jerez wrote:
>> > Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com> writes:
>> >=20
>> > > Enable HWP boost on Skylake server and workstations.
>> > >=20
>> >=20
>> > Please revert this series, it led to significant energy usage and
>> > graphics performance regressions [1].=20
>>=20
>> Which SKX platform is targeted to graphics?
> There are some Xeon E3, which is using SKL desktop CPUID.
> Do you have a SKL desktop with FADT pm profile=20
> acpi_gbl_FADT.preferred_profile =3D=3D PM_DESKTOP? You can take acpidump
> and check.
>
> Also what is
>  /sys/devices/system/cpu/intel_pstate/hwp_dynamic_boost
>

Eero is the one that reproduced the regressions -- Can you get Srinivas
this information from your system?

Thank you.

> Thanks,
> Srinivas
>
>>=20
>> >  The reasons are roughly the ones
>> > we discussed by e-mail off-list last April: This causes the
>> > intel_pstate
>> > driver to decrease the EPP to zero=20
>>=20
>> No. You didn't check this series. We are not using EPP at all.
>> The boost mechanism used here is not boost to max.
>>=20
>> Thanks,
>> Srinivas
>>=20
>> > when the workload blocks on IO
>> > frequently enough, which for the regressing benchmarks detailed in
>> > [1]
>> > is a symptom of the workload being heavily IO-bound, which means
>> > they
>> > won't benefit at all from the EPP boost since they aren't
>> > significantly
>> > CPU-bound, and they will suffer a decrease in parallelism due to
>> > the
>> > active CPU core using a larger fraction of the TDP in order to
>> > achieve
>> > the same work, causing the GPU to have a lower power budget
>> > available,
>> > leading to a decrease in system performance.
>> >=20
>> > You may want to give a shot to my previous suggestion of using [2]
>> > in
>> > order to detect whether the system is IO-bound, which you can use
>> > as
>> > an
>> > indicator that the optimization implemented in this series cannot
>> > possibly improve performance and can be expected to hurt energy
>> > usage.
>> >=20
>> > Thanks.
>> >=20
>> > [1] https://bugs.freedesktop.org/show_bug.cgi?id=3D107410
>> > [2] https://patchwork.kernel.org/patch/10312259/
>> >=20
>> > > Reported-by: Mel Gorman <mgorman@techsingularity.net>
>> > > Tested-by: Giovanni Gherdovich <ggherdovich@suse.cz>
>> > > Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.int
>> > > el
>> > > .com>
>> > > ---
>> > >  drivers/cpufreq/intel_pstate.c | 10 ++++++++++
>> > >  1 file changed, 10 insertions(+)
>> > >=20
>> > > diff --git a/drivers/cpufreq/intel_pstate.c
>> > > b/drivers/cpufreq/intel_pstate.c
>> > > index 70bf63bb4e0e..01c8da1f99db 100644
>> > > --- a/drivers/cpufreq/intel_pstate.c
>> > > +++ b/drivers/cpufreq/intel_pstate.c
>> > > @@ -1794,6 +1794,12 @@ static const struct x86_cpu_id
>> > > intel_pstate_cpu_ee_disable_ids[] =3D {
>> > >  	{}
>> > >  };
>> > >=20=20
>> > > +static const struct x86_cpu_id intel_pstate_hwp_boost_ids[]
>> > > __initconst =3D {
>> > > +	ICPU(INTEL_FAM6_SKYLAKE_X, core_funcs),
>> > > +	ICPU(INTEL_FAM6_SKYLAKE_DESKTOP, core_funcs),
>> > > +	{}
>> > > +};
>> > > +
>> > >  static int intel_pstate_init_cpu(unsigned int cpunum)
>> > >  {
>> > >  	struct cpudata *cpu;
>> > > @@ -1824,6 +1830,10 @@ static int intel_pstate_init_cpu(unsigned
>> > > int cpunum)
>> > >  			intel_pstate_disable_ee(cpunum);
>> > >=20=20
>> > >  		intel_pstate_hwp_enable(cpu);
>> > > +
>> > > +		id =3D x86_match_cpu(intel_pstate_hwp_boost_ids);
>> > > +		if (id)
>> > > +			hwp_boost =3D true;
>> > >  	}
>> > >=20=20
>> > >  	intel_pstate_get_cpu_pstates(cpu);
>> > > --=20
>> > > 2.13.6

--=-=-=--

--==-=-=
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iHUEAREIAB0WIQST8OekYz69PM20/4aDmTidfVK/WwUCW1zQ1gAKCRCDmTidfVK/
W2sCAP0d78328P9VrLs7stSaojcOxzy6M9e5dGzKty9yJ6ayEAD/QNqeTmpLM/D1
KSxbznmfTb2Qck9N6GFxDPiO+ZO8XzE=
=7Mbl
-----END PGP SIGNATURE-----
--==-=-=--