Hi, On 5/18/23 18:02, Andy Shevchenko wrote: > On Thu, May 18, 2023 at 6:32 PM Hans de Goede <hdegoede@xxxxxxxxxx> wrote: >> >> Hi All, >> >> Now that the atomisp driver supports v4l2-async sensor driver registration >> (I'll post this as a separate series), there is no longer a need to have >> atomisp specific sensor drivers and after cleanup atomisp sensor drivers >> can now be moved to drivers/media/i2c as regular v4l2 sensor drivers! > > Cool! > > But since the patch 9/9 in that series is WIP (as far as I can see), I > think we first need to concentrate to review that one. OTOH I'm not > sure I got this right that I'm thinking that this series depends on > that one one way or the other. Only patches 3/9 and 9/9 depend on the other series in the sense that they would break gc0310 support in atomisp without the other series. 3/9 can be moved to the end so the other 7 patches can be merged. Most important in this series and the reason for posting it separately is getting Sakari to ack patch 1/9 . Sakari, I know you don't want platform dependent code in the sensor drivers and I understand why. But in this case it is really unavoidable. To make this as painless as possible the helper function has been made so that all sensors which may be used on affected platforms only need this one extra line: v4l2_acpi_parse_sensor_gpios(&client->dev); And on none ACPI platforms or when client was not instantiated through ACPI this will be a stub / no-op. So I hope that you are ok with this as a solution for how the ACPI tables on these devices work ? Regards, Hans
