Re: [PATCH 18/18] i2c-cht-wc: Add device-properties for fusb302 integration

[Andy Shevchenko]

On Sun, Aug 6, 2017 at 6:05 PM, Hans de Goede <hdegoede@xxxxxxxxxx> wrote:
> On 06-08-17 16:35, Guenter Roeck wrote:
>> On 08/06/2017 05:35 AM, Hans de Goede wrote:

> FWIW all DSTDs I've seen are all copy and paste and all declare a INT33FE
> ACPI
> device with identical i2c client addresses which strongly suggests the
> use of the same combo. Note that on most devices this part of the DSTD is
> not active (_STA returns 0) because they actually use a different config.

Which is quite likely just blindly copied from a reference BIOS code.
(Reminds me that bug with GPIO pin setting for interrupt as output only)

> The only extra safe-guard I can come up with is a DMI string check, but that
> is sub-optimal since the DMI strings on these devices contain useful values
> as "Default String" still we could add it as an extra check.

...and in worst cases CPU model ID.

> Since I had the same concern I've done a web search and I've been unable
> to find any other devices which seem to use a Whiskey Cove PMIC, but that
> does not mean that there aren't any.


-- 
With Best Regards,
Andy Shevchenko