On Fri, 18 May 2018 18:07:26 +0200,
Pierre-Louis Bossart wrote:
>
> On 5/17/18 11:36 PM, Vinod wrote:
> > On 17-05-18, 17:53, Pierre-Louis Bossart wrote:
> >> HID is assumed to be made of TI PCI ID (0x104C) + part number, so all
> >> four 104C5121, 104C5122, 104C5141 104C5142 are valid.
> >>
> >> Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@xxxxxxxxxxxxxxx>
> >> ---
> >
> > ...
> >
> >> static struct i2c_driver pcm512x_i2c_driver = {
> >> .probe = pcm512x_i2c_probe,
> >> @@ -67,7 +81,8 @@ static struct i2c_driver pcm512x_i2c_driver = {
> >> .id_table = pcm512x_i2c_id,
> >> .driver = {
> >> .name = "pcm512x",
> >> - .of_match_table = pcm512x_of_match,
> >> + .of_match_table = of_match_ptr(pcm512x_of_match),
> >
> > this change may not be part of this and should be mentioned in the log
> >
> > Also quick grep tells me kernel has both :(
>
> I don't get your point. This is how we've added support for ACPI for
> every codec so far, and using those macros is the way to make sure the
> DT case still builds and isn't impacted.
Well, my interpretation is that this change is needed to allow the
driver built without CONFIG_OF, and it's not only about ACPI, per se.
Takashi
_______________________________________________
Alsa-devel mailing list
Alsa-devel@xxxxxxxxxxxxxxxx
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
