- 07 Oct, 2023 40 commits
-
-
Sakari Ailus authored
Split camera sensor documentation into user and kernel portions. This should make it easier for the user space developers to find the relevant documentation. Also add a list of exemplary drivers and add imx219 driver to it, besides those that were already mentioned. Signed-off-by:
Sakari Ailus <sakari.ailus@linux.intel.com> Reviewed-by:
Laurent Pinchart <laurent.pinchart@ideasonboard.com> Reviewed-by:
Jacopo Mondi <jacopo.mondi@ideasonboard.com> Signed-off-by:
Hans Verkuil <hverkuil-cisco@xs4all.nl>
-
Sakari Ailus authored
Add a definition for MIPI CSI-2 generic long packet types. The generic long packet types are numbered from 1 to 4. Signed-off-by:
-
Sakari Ailus authored
Check the validity of pad flags on entity init. Exactly one of the flags must be set. Signed-off-by:
-
Sakari Ailus authored
Print debug level information on returned frame descriptors. Signed-off-by:
-
Sakari Ailus authored
Clear frame descriptor before calling transmitter's get_frame_desc() op. Also remove the corresponding memset() calls from drivers. Signed-off-by:
Tomi Valkeinen <tomi.valkeinen@ideasonboard.com> Signed-off-by:
-
Sakari Ailus authored
With ipu bridge, endpoints may only be created when ipu bridge has initialised. This may happen after the sensor driver has first probed. Signed-off-by:
-
Sakari Ailus authored
Use sub-device active state. Rely on control handler lock to serialise access to the active state. Also clean up locking on s_stream handler. Signed-off-by:
-
Sakari Ailus authored
Enable runtime PM before registering the async subdev as the driver UAPI may become accessible immediately after the registration. Runtime PM needs to be enabled by that time. Signed-off-by:
-
Sakari Ailus authored
The s_stream is called to enable and to disable streaming on a sub-device. The caller may only call it to change the state, enabling streaming is not allowed when it is already disabled, and similarly for disabling streaming. Remove the check from the CCS driver. Signed-off-by:
-
Sakari Ailus authored
ccs_pm_get_init() depends on the return values > 0 of pm_runtime_get_sync(), thus it can't use pm_runtime_resume_and_get(). There's even a comment in the driver on this, a few lines above the code. Fixes: aa0adb399d09 ("media: i2c: Use pm_runtime_resume_and_get()") Signed-off-by: -
Sakari Ailus authored
Make use of sub-device active state. In most cases the effect on need for acquiring the mutex is non-existent as access to the driver's core data structure still needs to be serialised. This still removes a lot of code as the code paths for active and try state are the same in many cases. Signed-off-by:
-
Sakari Ailus authored
The available mbus codes will soon be needed earlier, at the time sub-devices are initialisaed. This is due to calling init_cfg() op via the v4l2_subdev_init_finalize(). Move ccs_get_mbus_formats() before ccs_init_subdev() calls. Signed-off-by:
-
Sakari Ailus authored
The media entity will soon need to be initialised before the sub-device init finalisation that allocates memory for sub-device state. Do that now. Signed-off-by:
-
Sakari Ailus authored
The ccs_create_subdev() function initialises a sub-device in the CCS driver, including CCS specific needs. Rename it as ccs_init_subdev() as it better reflects what the function does. Signed-off-by:
-
Sakari Ailus authored
Use init_cfg() instead of manually setting up defaults in file handle open. Signed-off-by:
-
Sakari Ailus authored
ccs_register_subdev() did not clean up the media entity in error case, do that now. Also switch to goto based error handling. Signed-off-by:
-
Sakari Ailus authored
Initialise the try sink compose rectangle size to the sink compose rectangle for binner and scaler sub-devices. This was missed due to the faulty condition that lead to the compose rectangles to be initialised for the pixel array sub-device where it is not relevant. Fixes: ccfc97bd ("[media] smiapp: Add driver") Cc: stable@vger.kernel.org Signed-off-by:
-
Sakari Ailus authored
Fix documentation for struct ccs_quirk, a device specific struct for managing deviations from the standard. The flags field was drifted away from where it should have been. Signed-off-by:
-
Sakari Ailus authored
Align lines for numbered list so that Sphinx produces an uniform output for all list entries. Also indent paragraphs of such list entries for consistency. Also use ReST numbered list syntax for the entries. Signed-off-by:
-
Laurent Pinchart authored
The exposure_max, exposure_def and hblank variables are only used in an inner scope in the imx219_set_pad_format() function. Move them to that scope to keep them closer to their usage and improve readability. Signed-off-by:
Dave Stevenson <dave.stevenson@raspberrypi.com> Signed-off-by:
-
Laurent Pinchart authored
Subdev state variables are named with a mix of 'state' and 'sd_state' through the driver. To improve consistency, name them all 'state'. Signed-off-by:
-
Laurent Pinchart authored
Calculate the crop rectangle size and location dynamically when setting the format, instead of storing it in the imx219_mode structure. This removes duplicated information from the mode, to guarantee consistency. Signed-off-by:
-
Laurent Pinchart authored
The IMX219 has distinct binning registers for the horizontal and vertical directions. Calculate their value and write them separately. Signed-off-by:
-
Laurent Pinchart authored
Instead of duplicating the logical implemented in the .set_fmt() operation in .init_cfg(), call .set_fmt() directly. This centralizes the format and crop rectangle calculations in a single place. Signed-off-by:
-
Laurent Pinchart authored
Simplify the imx219_set_stream() by removing an unneeded goto statement, and its corresponding error label. The natural flow of the function is correct. While at it, drop a useless comment. Signed-off-by:
-
Laurent Pinchart authored
Move functions around to group them by purpose, in order to improve readability. No functional change is intended. Signed-off-by:
-
Laurent Pinchart authored
The IMX219_VTS_* macros define default VTS values for the modes supported by the driver. They are used in a single place, and hinder readability compared to using the value directly as a decimal number. Drop them. Signed-off-by:
-
Laurent Pinchart authored
The mode field of the imx219 structure is only used in imx219_init_controls(), after the probe function sets it to point to the default mode. Use the default mode directly when initializing controls, and drop the mode field from the imx219 structure. Signed-off-by:
-
Laurent Pinchart authored
Use the active format height instead of the mode height in imx219_set_ctrl(). This prepares for dropping the mode field from the imx219 structure. The state is retrieved using v4l2_subdev_get_locked_active_state() as the subdev active state and the control handler share the same lock. Signed-off-by:
-
Laurent Pinchart authored
Compare the format and crop rectangle dimensions to infer binning settings, instead of storing the binning mode in the imx219_mode structure. This removes duplicate information from the mode. Signed-off-by:
-
Laurent Pinchart authored
Configure the crop-related registers from the values stored in the active crop rectangle instead of the mode structure. This removes usage of the mode from the imx219_set_framefmt(). No functional change is intended. Signed-off-by:
-
Laurent Pinchart authored
While the ycbcr_enc field doesn't apply to raw formats, leaving it uninitialized makes the driver behave in a less deterministic way. Fix it by picking the default value for the colorspace. Signed-off-by:
-
Laurent Pinchart authored
The imx219_set_binning() function sets registers based on the bpp value, which is computed in imx219_set_framefmt(). As both functions are called from the same place consecutively, and set registers based on the selected mode, merge them together to simplify the code. Signed-off-by:
-
Laurent Pinchart authored
Replace the per-mode register arrays with code that sets the same register values using the mode definitions. This avoids duplicating information in two different places. The error check for invalid formats in imx219_set_framefmt() is dropped as the format is guaranteed to be valid. Signed-off-by:
-
Laurent Pinchart authored
The 640x480 mode specifies incorrect values for the TP_WINDOW_WIDTH and TP_WINDOW_HEIGHT registers, which likely got copied from the 1640x1232 mode. They should be identical to the X_OUTPUT_SIZE and Y_OUTPUT_SIZE registers as for all the other modes, to avoid cropping the test pattern. Fix them. Signed-off-by:
-
Laurent Pinchart authored
The IMX219_REG_CSI_LANE_MODE is configured twice, once with a hardcoded value in the imx219_common_regs registers array, and once with the value appropriate for the system in imx219_configure_lanes(). The latter is enough, drop the former. Fixes: ceddfd44 ("media: i2c: imx219: Support four-lane operation") Suggested-by:
-
Laurent Pinchart authored
Define macros for all the known registers used in the register arrays, and use them to replace the numerical addresses. This improves readability. Signed-off-by:
-
Laurent Pinchart authored
Drop a handful of macros that are not used and don't provide any value. Signed-off-by:
-
Laurent Pinchart authored
Use the new common CCI register access helpers to replace the private register access helpers in the imx219 driver. This simplifies the driver by reducing the amount of code. Signed-off-by:
-
Laurent Pinchart authored
The MT9M114 is a CMOS camera sensor that combines a 1296x976 pixel array with a 10-bit dynamic range together with an internal ISP. The driver exposes two subdevs, one for the pixel array and one for the ISP (named IFP for Image Flow Processor). Major supported features are - Full configuration of analog crop and binning in the pixel array - Full configuration of scaling in the ISP - Automatic exposure and white balance - Manual exposure and analog gain - Horizontal and vertical flip Signed-off-by:
-
