Re: [PATCH v5 2/3] drm/i915: Lookup and attach ACPI device node for connectors

[Rajat Jain]

Hi Jani,

Thank you for the review. Please see inline.

On Fri, Jan 24, 2020 at 3:37 AM Jani Nikula <jani.nikula@xxxxxxxxxxxxxxx> wrote:
>
> On Fri, 20 Dec 2019, Rajat Jain <rajatja@xxxxxxxxxx> wrote:
> > Lookup and attach ACPI nodes for intel connectors. The lookup is done
> > in compliance with ACPI Spec 6.3
> > https://uefi.org/sites/default/files/resources/ACPI_6_3_final_Jan30.pdf
> > (Ref: Pages 1119 - 1123).
> >
> > This can be useful for any connector specific platform properties. (This
> > will be used for privacy screen in next patch).
> >
> > Signed-off-by: Rajat Jain <rajatja@xxxxxxxxxx>
> > ---
> > v5: same as v4
> > v4: Same as v3
> > v3: fold the code into existing acpi_device_id_update() function
> > v2: formed by splitting the original patch into ACPI lookup, and privacy
> >     screen property. Also move it into i915 now that I found existing code
> >     in i915 that can be re-used.
> >
> >  drivers/gpu/drm/i915/display/intel_acpi.c     | 24 +++++++++++++++++++
> >  .../drm/i915/display/intel_display_types.h    |  3 +++
> >  drivers/gpu/drm/i915/display/intel_dp.c       |  3 +++
> >  3 files changed, 30 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/i915/display/intel_acpi.c b/drivers/gpu/drm/i915/display/intel_acpi.c
> > index e21fb14d5e07..101a56c08996 100644
> > --- a/drivers/gpu/drm/i915/display/intel_acpi.c
> > +++ b/drivers/gpu/drm/i915/display/intel_acpi.c
> > @@ -222,11 +222,23 @@ static u32 acpi_display_type(struct intel_connector *connector)
> >       return display_type;
> >  }
> >
> > +/*
> > + * Ref: ACPI Spec 6.3
> > + * https://uefi.org/sites/default/files/resources/ACPI_6_3_final_Jan30.pdf
> > + * Pages 1119 - 1123 describe, what I believe, a standard way of
> > + * identifying / addressing "display panels" in the ACPI. It provides
> > + * a way for the ACPI to define devices for the display panels attached
> > + * to the system. It thus provides a way for the BIOS to export any panel
> > + * specific properties to the system via ACPI (like device trees).
> > + */
> >  void intel_acpi_device_id_update(struct drm_i915_private *dev_priv)
> >  {
> >       struct drm_device *drm_dev = &dev_priv->drm;
> >       struct intel_connector *connector;
> >       struct drm_connector_list_iter conn_iter;
> > +     struct device *dev = &drm_dev->pdev->dev;
>
> Hmm, already pushed patch 1 with the unfortunate "drm_dev" local. We use
> "dev" for struct drm_device * and almost never use struct device.

Sorry, I did not know. I'll send an independent fixup patch for patch
1 on top of drm-intel-next-queued (or let me know if you want to
handle it). I will also change this patch to rename the variable.

>
> > +     struct acpi_device *conn_dev;
> > +     u64 conn_addr;
> >       u8 display_index[16] = {};
> >
> >       /* Populate the ACPI IDs for all connectors for a given drm_device */
> > @@ -242,6 +254,18 @@ void intel_acpi_device_id_update(struct drm_i915_private *dev_priv)
> >               device_id |= display_index[type]++ << ACPI_DISPLAY_INDEX_SHIFT;
> >
> >               connector->acpi_device_id = device_id;
> > +
> > +             /* Build the _ADR to look for */
> > +             conn_addr = device_id | ACPI_DEVICE_ID_SCHEME |
> > +                             ACPI_BIOS_CAN_DETECT;
> > +
> > +             DRM_DEV_INFO(dev, "Checking connector ACPI node at _ADR=%llX\n",
> > +                          conn_addr);
>
> This is more than a little verbose. One line of INFO level dmesg for
> every connector at boot and at resume.
>
> Please use the new drm_dbg_kms() macro for this.

Will do.

>
> > +
> > +             /* Look up the connector device, under the PCI device */
> > +             conn_dev = acpi_find_child_device(ACPI_COMPANION(dev),
> > +                                               conn_addr, false);
> > +             connector->acpi_handle = conn_dev ? conn_dev->handle : NULL;
>
> acpi_device_handle(conn_dev)
>

Will do.


> >       }
> >       drm_connector_list_iter_end(&conn_iter);
> >  }
> > diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
> > index 1a7334dbe802..0a4a04116091 100644
> > --- a/drivers/gpu/drm/i915/display/intel_display_types.h
> > +++ b/drivers/gpu/drm/i915/display/intel_display_types.h
> > @@ -407,6 +407,9 @@ struct intel_connector {
> >       /* ACPI device id for ACPI and driver cooperation */
> >       u32 acpi_device_id;
> >
> > +     /* ACPI handle corresponding to this connector display, if found */
> > +     void *acpi_handle;
> > +
>
> The type is acpi_handle. It's none of our business to know what the
> underlying type is.

Will do.

>
> >       /* Reads out the current hw, returning true if the connector is enabled
> >        * and active (i.e. dpms ON state). */
> >       bool (*get_hw_state)(struct intel_connector *);
> > diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
> > index b05b2191b919..93cece8e2516 100644
> > --- a/drivers/gpu/drm/i915/display/intel_dp.c
> > +++ b/drivers/gpu/drm/i915/display/intel_dp.c
> > @@ -45,6 +45,7 @@
> >  #include "i915_debugfs.h"
> >  #include "i915_drv.h"
> >  #include "i915_trace.h"
> > +#include "intel_acpi.h"
> >  #include "intel_atomic.h"
> >  #include "intel_audio.h"
> >  #include "intel_connector.h"
> > @@ -6623,6 +6624,8 @@ intel_dp_add_properties(struct intel_dp *intel_dp, struct drm_connector *connect
> >
> >               connector->state->scaling_mode = DRM_MODE_SCALE_ASPECT;
> >
> > +             /* Lookup the ACPI node corresponding to the connector */
> > +             intel_acpi_device_id_update(dev_priv);
>
> Auch, this is problematic. It iterates all connectors, for every DP
> connector being added. In the middle of registering all connectors.
>
> From the POV of this patch alone, this is also unnecessary. This gets
> called via intel_opregion_register() after all connectors have been
> registered.
>
> I am aware it's not enough for your next patch, because it will need the
> acpi handle right here.
>
> I'm wondering if we need to maintain display_index[] in struct
> drm_i915_private, and update that as connectors get added instead of all
> at once in the end.

Sure, I can do that.

> connector->acpi_device_id never changes, does it,
> even though we keep updating it?

This is the part I am not so sure about - I hypothesized that theory
because of the current behavior in code (i.e. it is getting updated in
intel_opregion_resume() path). May be it does not change, I was not
sure, so I did not want to create any regression in the intel_opregion
code that I did not understand. I tried on my system and as far as I
could experiment, I did not see it changing. Please let me know if you
would like me to change my code to:

1) Maintain drm_i915_private->display_index[] and update it as
connectors are added.
2) Remove the code to update it on every resume and while registering
the connector.

Thanks & Best Regards,

Rajat

>
> BR,
> Jani.
>
>
> >       }
> >  }
>
> --
> Jani Nikula, Intel Open Source Graphics Center