Re: [PATCH v3 0/2] Add IMX219 CMOS image sensor support

Andrey Konovalov <andrey.konovalov@xxxxxxxxxx> · Sat, 11 Jan 2020 20:46:33 +0300

Hi Ezequiel,

On 11.01.2020 07:45, Ezequiel Garcia wrote:
Hello Andrey,

Thanks for submitting a new version.

On Fri, 2020-01-10 at 23:09 +0300, Andrey Konovalov wrote:
This patchset adds support for IMX219 CMOS image sensor from Sony.
Sony IMX219 is an 8MPix, 1/4.0-inch CMOS active pixel digital image sensor
with an active array size of 3280H x 2464V. It is programmable through
I2C interface. Image data are sent through MIPI CSI-2, which can be configured
as either 2 or 4 data lanes, but this driver currently only supports 2 lanes.
The currently supported resolutions are 3280x2464 @ 15fps, 1920x1080 @ 30fps
(cropped FOV), and 1640x1232 (2x2 binned) @ 30fps.

The driver has been tested with Raspberry Pi Camera Module v2 connected to
Raspberry Pi Zero W.

Changes since v2 [1]:

dt-bindings:
   - "clock-names" property removed
   - "xclr-gpios" property renamed to "reset-gpios"
   - the camera-clk mode moved out of sensor device node
   - "clock-lanes" property removed (the sensor doesn't support lane reordering)
   - "clock-noncontinuous" description made more clear (thanks Sakari)
   - "data-lanes" property reworked: it is now optional, and if it is not
     present the driver should assume four-lane operation. For two-lane
     operation (the only mode supported by the current driver) this property
     must be present and set to <1 2>

imx219 sensor driver:
   - "xclr-gpios" property renamed to "reset-gpios", the corresponding
     struct gpio_desc field in the imx219 structure is renamed to reset_gpio
   - in the driver probe() a test to check that the number of CSI-2 data lanes
     is supported by the driver was added
   - devm_clk_get() is now called with NULL as the 2nd argument, as there is
     just single clock, and there is no need to use clock ID
   - error messages are added when the driver fails to get regulators, to init
     media entity pads, or to register sensor sub-device

It seems you missed Sakari's review comments on v2,
see https://patchwork.linuxtv.org/patch/60925/#114407

Hm.. Indeed. I missed Sakari's comments on the imx219.c v2 somehow.
I'll post an updated patchset shortly.

Thanks!
Andrey

Thanks!
Eze