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;
Message-ID: <fa84b0098551123f34a68b3c0d9c7aa12f592bbf.camel@linux.intel.com>
Subject: Re: [PATCH V2 2/2] platform/x86/intel/pmc/mtl: Put devices in D3
 during resume
From:   "David E. Box" <david.e.box@linux.intel.com>
Reply-To: david.e.box@linux.intel.com
To:     Hans de Goede <hdegoede@redhat.com>, linux-kernel@vger.kernel.org,
        platform-driver-x86@vger.kernel.org
Cc:     markgross@kernel.org, irenic.rajneesh@gmail.com,
        ilpo.jarvinen@linux.intel.com, xi.pardee@intel.com,
        rajvi.jingar@linux.intel.com
Date:   Mon, 12 Jun 2023 11:02:38 -0700
In-Reply-To: <e5fce103-fbb6-0eb6-f6ff-4bfeadd89c90@redhat.com>
References: <20230607233849.239047-1-david.e.box@linux.intel.com>
         <20230607233849.239047-2-david.e.box@linux.intel.com>
         <e5fce103-fbb6-0eb6-f6ff-4bfeadd89c90@redhat.com>
Organization: David E. Box
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.44.4 (3.44.4-3.fc36) 
MIME-Version: 1.0
Precedence: bulk

Hi Hans,

On Mon, 2023-06-12 at 11:42 +0200, Hans de Goede wrote:
> Hi David,
>=20
> On 6/8/23 01:38, David E. Box wrote:
> > An earlier commit placed some driverless devices in D3 during boot so t=
hat
> > they don't block package cstate entry on Meteor Lake. Also place these
> > devices in D3 after resume from suspend.
> >=20
> > Fixes: 336ba968d3e3 ("platform/x86/intel/pmc/mtl: Put GNA/IPU/VPU devic=
es in
> > D3")
> > Signed-off-by: David E. Box <david.e.box@linux.intel.com>
>=20
> Thank you for your patch.
>=20
> There is one thing which has me worried here:
>=20
> What about when real proper drivers show up for these blocks?
>=20
> I know that at least some people will likely be using the out of tree IPU=
6
> driver with the IPU block.
>=20
> And having 2 different drivers poke at the hw state seems like a bad idea=
 to
> me.
>=20
> Maybe we can add a check if no driver is bound and only set the state to =
D3 if
> no driver is bound?

This check exists but is not shown in the patch. mtl_set_device_d3() gets t=
he
device lock and checks to see if dev.driver is NULL before putting in D3. T=
his
was checked with the GNA driver installed.

David

>=20
> Regards,
>=20
> Hans
>=20
>=20
>=20
> > ---
> >=20
> > V2 - rename mtl_fixup to mtl_d3_fixup. Call it from new mtl_resume
> > =C2=A0=C2=A0=C2=A0=C2=A0 function, followed by the common resume. Sugge=
sted by Ilpo.
> >=20
> > =C2=A0drivers/platform/x86/intel/pmc/mtl.c | 29 ++++++++++++++++++++---=
-----
> > =C2=A01 file changed, 21 insertions(+), 8 deletions(-)
> >=20
> > diff --git a/drivers/platform/x86/intel/pmc/mtl.c
> > b/drivers/platform/x86/intel/pmc/mtl.c
> > index e8cc156412ce..2b00ad9da621 100644
> > --- a/drivers/platform/x86/intel/pmc/mtl.c
> > +++ b/drivers/platform/x86/intel/pmc/mtl.c
> > @@ -68,16 +68,29 @@ static void mtl_set_device_d3(unsigned int device)
> > =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0}
> > =C2=A0}
> > =C2=A0
> > -void mtl_core_init(struct pmc_dev *pmcdev)
> > +/*
> > + * Set power state of select devices that do not have drivers to D3
> > + * so that they do not block Package C entry.
> > + */
> > +static void mtl_d3_fixup(void)
> > =C2=A0{
> > -=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0pmcdev->map =3D &mtl_reg_map=
;
> > -=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0pmcdev->core_configure =3D m=
tl_core_configure;
> > -
> > -=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0/*
> > -=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 * Set power state of select=
 devices that do not have drivers to D3
> > -=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 * so that they do not block=
 Package C entry.
> > -=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 */
> > =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0mtl_set_device_d3(MTL_G=
NA_PCI_DEV);
> > =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0mtl_set_device_d3(MTL_I=
PU_PCI_DEV);
> > =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0mtl_set_device_d3(MTL_V=
PU_PCI_DEV);
> > =C2=A0}
> > +
> > +static int mtl_resume(struct pmc_dev *pmcdev)
> > +{
> > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0mtl_d3_fixup();
> > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0return pmc_core_resume_commo=
n(pmcdev);
> > +}
> > +
> > +void mtl_core_init(struct pmc_dev *pmcdev)
> > +{
> > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0pmcdev->map =3D &mtl_reg_map=
;
> > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0pmcdev->core_configure =3D m=
tl_core_configure;
> > +
> > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0mtl_d3_fixup();
> > +
> > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0pmcdev->resume =3D mtl_resum=
e;
> > +}
>=20