Re: [PATCH 1/2] ACPI: x86: Add skip i2c clients quirk for Medion Lifetab S10346

[Hans de Goede <hdegoede@redhat.com>]

Hi,

On 12/8/22 14:00, Andy Shevchenko wrote:
> On Thu, Dec 8, 2022 at 1:02 PM Hans de Goede <hdegoede@redhat.com> wrote:
>>
>> The Medion Lifetab S10346 is a x86 tablet which ships with Android x86 as
>> factory OS. The Android x86 kernel fork ignores I2C devices described in
>> the DSDT, except for the PMIC and Audio codecs.
>>
>> As usual the Medion Lifetab S10346's DSDT contains a bunch of extra I2C
>> devices which are not actually there, causing various resource conflicts.
>> Add an ACPI_QUIRK_SKIP_I2C_CLIENTS quirk for the Medion Lifetab S10346 to
>> the acpi_quirk_skip_dmi_ids table to woraround this.
> 
> workaround
> 
> 
> Both look good to me,
> Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>

Thanks. I have merged 2/2 into my review-hans branch now.

> Side note. We adding a lot of DMI matching information here and there
> in the drivers and often it gets duplicated in handful of times, Maybe
> at some point we can create a global enum and matching table, let's
> say driver/platform/x86/dmi-platforms.c with
> include/linux/platform_data/x86/platform.h with global enum of the
> platforms?

Maybe, I'm not sure atm the table in x86-android-tablets.c is
__initconst so it gets free-ed. So resident code-size wise
it would not matter.

And it would require replacing the dmi_system_id table
in x86-android-tablets.c with another way to map the enum
values (replacing the DMI matches) to driver_data for
the specific models. Which at best would also be an
__initconst mapping table...

Regards,

Hans