Re: [PATCH v3 2/4] ACPI / scan: Create platform device for fwnodes with multiple i2c devices

From: Hans de Goede
Date: Wed Aug 08 2018 - 04:07:35 EST


Hi,

On 07-08-18 13:49, Andy Shevchenko wrote:
On Tue, 2018-08-07 at 13:29 +0200, Hans de Goede wrote:
Hi,

On 07-08-18 13:19, Andy Shevchenko wrote:
On Tue, 2018-08-07 at 10:05 +0200, Hans de Goede wrote:

+ /*
+ * These devices have multiple I2cSerialBus resources and an
i2c-client
+ * must be instantiated for each, each with its own
i2c_device_id.
+ * Normally we only instantiate an i2c-client for the first
resource,
+ * using the ACPI HID as id. These special cases are handled by
the
+ * drivers/platform/x86/i2c-multi-instantiate.c driver, which
knows
+ * which i2c_device_id to use for each resource.
+ */
+ static const struct acpi_device_id i2c_multi_instantiate_ids[] =
{
+ {"BSG1160", 0},
+ {"", 0},
+ };

Style nits:
- can we move it outside of function?

Sure, but there are 2 existing users of an array of acpi_device_id-s
combined with an acpi_match_device_ids() call and both have the array
inside the function, so for consistency it seems better to keep it
where it is.

Hmm... OK.

- is this existing style in the file and / or files in this folder
for
IDs? (I mean unnecessary 0:s and empty string?

It seems that all variants one can come up with are already used
inside
this single file.

Ah, that's sad.

I agree that less is more, so I will change this to:

static const struct acpi_device_id
i2c_multi_instantiate_ids[] = {

{"BSG1160", },
{}
};

In case if it mimics already existing style, looks quite good to me
(otherwise perhaps comma inside {} can also be removed).


For v4.

Does it make sense to test v3 on your opinion? Or better to wait for v4?

Sorry for being a bit slow to answer, I'm about to send out v4, so probably
best to wait for that now. Note the 2 will be functionally identical,
I mainly fixed / clarified commit messages and the MAINTAINERS entry +
the small style fixed discussed above.

Regards,

Hans