[Intel-gfx] [PATCH 3/7] drm/i915/dg2: DG2 MBD config
Matt Roper
matthew.d.roper at intel.com
Tue May 31 15:39:51 UTC 2022
On Sun, May 29, 2022 at 09:44:38PM -0700, Gupta, Anshuman wrote:
>
>
> > -----Original Message-----
> > From: Nikula, Jani <jani.nikula at intel.com>
> > Sent: Thursday, May 19, 2022 2:57 PM
> > To: Gupta, Anshuman <anshuman.gupta at intel.com>; intel-
> > gfx at lists.freedesktop.org
> > Cc: Nilawar, Badal <badal.nilawar at intel.com>; Ewins, Jon
> > <jon.ewins at intel.com>; Vivi, Rodrigo <rodrigo.vivi at intel.com>; Deak, Imre
> > <imre.deak at intel.com>; Gupta, Anshuman <anshuman.gupta at intel.com>
> > Subject: Re: [PATCH 3/7] drm/i915/dg2: DG2 MBD config
> >
> > On Wed, 18 May 2022, Anshuman Gupta <anshuman.gupta at intel.com> wrote:
> > > Add DG2 Motherboard Down Config check support.
> > >
> > > BSpec: 44477
> > > Cc: Rodrigo Vivi <rodrigo.vivi at intel.com>
> > > Signed-off-by: Anshuman Gupta <anshuman.gupta at intel.com>
> > > ---
> > > drivers/gpu/drm/i915/display/intel_opregion.c | 2 ++
> > > drivers/gpu/drm/i915/i915_drv.h | 9 +++++++++
> > > 2 files changed, 11 insertions(+)
> > >
> > > diff --git a/drivers/gpu/drm/i915/display/intel_opregion.c
> > > b/drivers/gpu/drm/i915/display/intel_opregion.c
> > > index 3dcd54517b89..dec7628522c5 100644
> > > --- a/drivers/gpu/drm/i915/display/intel_opregion.c
> > > +++ b/drivers/gpu/drm/i915/display/intel_opregion.c
> > > @@ -1271,6 +1271,8 @@ intel_opregion_vram_sr_required(struct
> > > drm_i915_private *i915)
> > >
> > > if (IS_DG1(i915))
> > > return intel_opregion_dg1_mbd_config(i915);
> > > + else if (IS_DG2_MBD(i915))
> > > + return true;
> > >
> > > return false;
> > > }
> > > diff --git a/drivers/gpu/drm/i915/i915_drv.h
> > > b/drivers/gpu/drm/i915/i915_drv.h index 10f273800645..c5ecc490dced
> > > 100644
> > > --- a/drivers/gpu/drm/i915/i915_drv.h
> > > +++ b/drivers/gpu/drm/i915/i915_drv.h
> > > @@ -1071,6 +1071,15 @@ IS_SUBPLATFORM(const struct drm_i915_private
> > *i915,
> > > IS_SUBPLATFORM(dev_priv, INTEL_DG2, INTEL_SUBPLATFORM_G11)
> > #define
> > > IS_DG2_G12(dev_priv) \
> > > IS_SUBPLATFORM(dev_priv, INTEL_DG2, INTEL_SUBPLATFORM_G12)
> > > +/*
> > > + * FIXME: Need to define a new SUBPLATFORM
> > INTEL_SUBPLATFORM_DG2_MBD
> > > + * for PCI id range 5690..5695, but G10, G11 SUBPLATFROM conflicts
> > > + * with those pci id range.
> > > + */
> > > +#define DG2_MBD_CONFIG_MASK GENMASK(7, 4)
> > > +#define DG2_MBD_CONFIG_VAL
> > FIELD_PREP(DG2_MBD_CONFIG_MASK, 9)
> > > +#define IS_DG2_MBD(dev_priv) (IS_DG2(dev_priv) && \
> > > + (INTEL_DEVID(dev_priv) &
> > DG2_MBD_CONFIG_MASK) ==
> > > +DG2_MBD_CONFIG_VAL)
> >
> > No. Please don't do *any* magic masking or comparison of PCI ID bits or
> > bitfields.
> Hi Matt,
> We need to distinguish between DG2 NB MBD and rest such that i915 can
> figure out opregion vram_sr is requires for the DG2 platform. Could
> you please suggestion on that.
If you truly need to distinguish a specific sub-type of DG2, then
creating a proper subplatform (with its own list of PCI IDs, matched in
intel_device_info_subplatform_init() is the way to go. As far as I
recall, it should be okay to have multiple subplatform bits set for a
platform, so for example you can have both the G11 and a MBD subplatform
bits set on a device and they can each be tested independently without
interfering with each other.
I'm not too familiar with the feature you're working on here. Is there
a way we can detect whether it's supported by querying the pcode? Or
what happens if you send your pcode request on a platform that doesn't
support it? Do you just get a regular error back so that the driver
would know to give up and move on, or would it actually cause some kind
of behavioral problem?
Matt
> Thanks,
> Anshuman Gupta.
> >
> > BR,
> > Jani.
> >
> > > #define IS_ADLS_RPLS(dev_priv) \
> > > IS_SUBPLATFORM(dev_priv, INTEL_ALDERLAKE_S,
> > INTEL_SUBPLATFORM_RPL)
> > > #define IS_ADLP_N(dev_priv) \
> >
> > --
> > Jani Nikula, Intel Open Source Graphics Center
--
Matt Roper
Graphics Software Engineer
VTT-OSGC Platform Enablement
Intel Corporation
More information about the Intel-gfx
mailing list