On Wed, 2017-03-08 at 18:05 +0100, Hans de Goede wrote:

Hi,

On 08-03-17 12:46, Andy Shevchenko wrote:

quoted

On Wed, 2017-03-08 at 12:27 +0100, Hans de Goede wrote:

quoted

On 08-03-17 11:30, Andy Shevchenko wrote:

quoted

On Wed, 2017-03-08 at 10:08 +0100, Hans de Goede wrote:

quoted

On 07-03-17 14:55, Hans de Goede wrote:

quoted

NULL sounds to me a bit clearer in this case, since the original
name of
connection IDs with underscores, not dashes.

Ok, so pass NULL and then drop the patch to add the mapping table,
because with a NULL con-id that won't be necessary right ?

I've just given this a spin (patch to pass NULl attached), your
patch to add the GPIO ACPI mapping table dropped and this works well
I agree just passing NULL as con-id is the better solution for
soc_button_array.

Yeah. The attached patch you sent is fine by me.

quoted

quoted

quoted

quoted

I think that "extcon: int3496: Add GPIO ACPI mapping table"
will
need
a similar change (I haven't tested it yet).

So I assume you want to do the same (pass NULL as con-id to
gpiod_get_index()) for the extcon-in3496 driver or do you want
to keep the GPIO ACPI mapping table there?

A slightly preferable table variant (needs to be fixed I guess) because
initial one used to have different labels.

Though if you would like to move to NULL, I wouldn't object.

-- 
Andy Shevchenko [off-list ref]
Intel Finland Oy