Age | Commit message (Collapse) | Author |
|
This new pipeline handler aims at supporting any simple device without
requiring any device-specific code. Simple devices are currently defined
as a graph made of one or multiple camera sensors and a single video
node, with each sensor connected to the video node through a linear
pipeline.
The simple pipeline handler will automatically parse the media graph,
enumerate sensors, build supported stream configurations, and configure
the pipeline, without any device-specific knowledge. It doesn't support
configuration of any processing in the pipeline at the moment, but may
be extended to support simple processing such as format conversion or
scaling in the future.
The only device-specific information in the pipeline handler is the list
of supported drivers, required for device matching. We may be able to
remove this in the future by matching with the simple pipeline handler
as a last resort option, after all other pipeline handlers have been
tried.
Signed-off-by: Martijn Braam <martijn@brixit.nl>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Andrey Konovalov <andrey.konovalov@linaro.org>
|
|
Add support for the recent V4L2 extension to VIDIOC_ENUM_FMT that allows
filtering pixel formats by media bus codes.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
This integrates the Linux kernel upstream commits e5b6b07a1b45 ("media:
v4l2: Extend VIDIOC_ENUM_FMT to support MC-centric devices") and
f645e6256bd1 ("media: v4l2-dev/ioctl: Add V4L2_CAP_IO_MC") that are
scheduled for v5.8.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
Should be $includedir/libcamera, as we install into a subdirectory name
'libcamera'. This is fixed by passing the 'subdirs' argument to the pkgconfig
generator.
Signed-off-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
The camera3_hal.cpp is incorrectly licensed as GPL-2.0-or-later, instead
of LGPL-2.1-or-later as the rest of the HAL implementation. Fix it.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Jacopo Mondi <jacopo@jmondi.org>
|
|
The EXIF IFD is incorrectly chained to IFD 0 in addition to being a
referenced as a sub IFD through the EXIFIFD tag. While the libtiff API
doesn't clearly document why this happens, inspection of the
TIFFWriteDirectory() source code show that the function treats the IFD
being written as containing an image, which isn't correct for the EXIF
IFD. Use TIFFWriteCustomDirectory() instead, which fixes the problem.
The resulting DNG file can now be opened with darktable in addition to
rawtherapee.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
|
|
The CameraSensor class assumes that camera sensors support the exact
same list of sizes of all media bus codes. While allowing a simpler API,
this assumption is incorrect and is blocking usage of some camera
sensors.
Relaxing the constraint is possible without changes to the CameraSensor
API syntax, but requires changing its semantics. The sizes() function
now returns the list of all sizes for all media bus codes, and the
getFormat() function now searches in all supported media bus codes. The
former is likely not the most useful option for pipeline handlers, but
the sizes() function is currently unused. Designing a better API will
require inspecting current and expected future use cases in pipeline
handlers to determine proper heuristics.
While at it, fix a small typo in an unrelated comment.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
Todo statements should not end with a colon, remove it.
Signed-off-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
While the DNG specification supports greyscale ("BlackIsZero") for
thumbnails, RawTherapee seems to have trouble reading them. Generate
thumbnails in RGB instead (but still with greyscale content, to avoid
having to interpolate colour components).
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
Populate the DNG black level, ISO speed rating and exposure time from
metadata. The ISO speed rating and exposure time are standardized as
EXIF tags, not TIFF tags, and require a separate EXIF IFD.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
Generate a greyscale, 1/16 resolution thumbnail and add it to the DNG
file. This requires shuffling the RAW image generation as the thumbnail
has to be stored in the main IFD as per the DNG and TIFF/EP
specifications.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
Name arguments to the FormatInfo::packScanline function pointer to make
it easier to understand its usage when reading the function declaration.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
The DNG writer will use the metadata to populate DNG tags.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
While most camera sensors are exposed to userspace as a single subdev,
this is not always the case. A notable exception is the smiapp driver,
which exposes two or three subdevs. The subdev at the output of the
camera sensor will thus have more than one pad. Prepare for this by
removing the single-pad restriction, and locating the source pad
dynamically.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
Avoid the need for a manual delete in the destructor by using a unique
pointer.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
The CameraSensor::logPrefix() function accesses the entity from the
subdev. Remove the level of indirection as we stored a pointer to the
entity internally.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
While reformatting the table the field name was missed for one entry,
add it.
Fixes: db7235b7141aa4e2 ("qcam: Add DNGWriter")
Signed-off-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
Add a toolbar button that captures RAW data to disk. The button is only
enabled if the camera is configured to provide a raw stream to the
application.
Only when the capture action is triggered will a request with a raw
buffer be queued to the camera.
Signed-off-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
Add an initial DNG file writer. The writer can only deal with a small
set of pixel formats. The generated file is consumable by standard
tools. The writer needs to be extended to write more metadata to the
generated file.
Signed-off-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
Allow a second stream to be configured for raw capture. This change only
adds support for configuring and allocating buffers for the second
stream. Later changes are needed to queue the allocated buffers to the
camera when the user wishes to capture a raw frame.
Signed-off-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
Since the Linux kernel commit:
69e39d40587b ("media: vimc: Implement get/set selection in sink")
the crop rectangle on the VIMC scaler sink pad needs to be
reset to match the requested format to avoid hitting a pipeline format
misconfiguration when the applied format is larger than the one
set in a previous capture session.
As the V4L2 specification is not clear on what the correct behaviour
is, if the crop rectangle should be reset automatically or not on a
set_fmt operation, momentary fix the pipeline handler to please the
driver and manually reset the crop rectangle.
Signed-off-by: Jacopo Mondi <jacopo@jmondi.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
Store the MediaDevice for the VIMC instance in the VimcCameraData class.
This will be used to check the media device version at runtime.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
Add a MediaDevice::version() function to expose the media device API
version number. This can be useful to support driver API changes in
pipeline handlers.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
Expose the image stride which may be retrieved after a video device
has been configured. It may only be retrieved at that point as the
assignment of video devices takes place at this point.
In the future video devices should be assigned at configuration
validation time and the stride value retrieved at that point.
Signed-off-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
To prepare for more widespread use of Python in libcamera, ignore Python
cache and compiled files produced when importing modules directly from
the source tree.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
gcc 7.1 introduced processor-specific ABI breakages related to parameter
passing on ARM platforms. This generates a large number of messages
during compilation with gcc >=7.1 until gcc 9. Silence them.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
|
|
If the camera can not generate a configuration from the requested roles
it returns a nullptr which leads to a nullptr dereference. Fix this by
adding a check that the camera generated a configuration before trying
to access it.
Signed-off-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
Use the StreamKeyValueParser helper to parse stream configuration from
the command line. This extends qcam to accept role hints and pixel
format in addition to a size.
Currently only one viewfinder stream is supported, add a check to keep
this behavior. Going forward this restriction will be lifted to support
more then one stream.
Signed-off-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
Use the StreamOptionsParser helper to parse stream configuration from
the command line.
Signed-off-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
Create a new helper class StreamKeyValueParser to parse command line
options describing stream configurations. The goal is to share this new
class between cam and qcam.
Signed-off-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
Extend OptionsBase<T> with a public invalidate() method. This allows for
further examination of the options and if found unsuitable be
invalidated. The intended user for this new interface are subclasses of
KeyValueParser.
Signed-off-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
Prepare for sub-classing of the KeyValueParser by making the parse()
method virtual.
Signed-off-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
When one of the camera is selected and opened from the
"Select Cameras" items list, the entry of the combo-box
in the main-window doesn't update its item index to reflect
the camera which was earlier selected. Fix that.
Signed-off-by: Umang Jain <email@uajain.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Tested-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
This commit introduces no functional changes.
This is required so that the combo-box list can be managed
conveniently from various private functions in subsequent
commit.
Signed-off-by: Umang Jain <email@uajain.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
The V4L2_PIX_FMT_SBGGR10P format was misspecified as
V4L2_PIX_FMT_SBGGR10. Fix it.
Fixes: 00d639d951e7 ("libcamera: formats: Add support for 8-, 10- and 12-bit Bayer formats")
Reported-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Tested-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
|
|
Add three additional fields to PixelFormatInfo to report the number of
bits per pixel, the colour encoding, and the data packing.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
|
|
Add the 8-, 10- and 12-but Bayer formats, in both unpacked and
MIPI-packed variants, to the format tables in PixelFormatInfo and
V4L2PixelFormat.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
|
|
To prepare for storing more information about pixel formats in
PixelFormatInfo, move the class to formats.cpp and document it. The
pixel formats database is moved to the same file, and a new static
function is added to PixelFormatInfo to retrieve a PixelFormatInfo for a
PixelFormat.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
|
|
Replace the two big manual switch...case with maps. This should not just
improve efficiency when we will have a larger number of formats, but
also paves the way to storing more format information to create
additional helpers.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
|
|
Move the DRM/V4L2 format conversion code from V4L2VideoDevice to
V4L2PixelFormat. This is a more natural home for the code.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
|
|
Move the V4L2PixelFormat class to a new file to prepare for additional
changes that will make it grow. No functional modification is included.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
Commit 7206035ee609 ("libcamera: Regenerate IPA module signatures at
install time") replaced installation of the IPA module signatures with
an install script that signs all modules. While doing so, it
inadvertently also disabled generation of the signature at build time by
default. This breaks running libcamera binaries from the build
directory. Fix it.
Fixes: 7206035ee609 ("libcamera: Regenerate IPA module signatures at install time")
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
Call realpath() to strip out any levels of indirection required in
referencing the root build directory.
This simplifies the debug output when reporting and parsing paths.
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
The comments describe functionality before libcameraBuildPath() was
adapted from libcameraPath() to return the root of the build, and
described having to take a relative path.
This is no longer the case, and the comment is inaccurate. Fix it.
Fixes: 1b8ac8473c94 ("libcamera: utils: Adapt libcameraPath to match use cases")
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
Make the IPAModule class inherit from Loggable to log the IPA module
name in all messages.
Before:
[14:19:55.743795676] [20250] ERROR IPAModule ipa_module.cpp:279 IPA module is not an ELF file
After:
[14:20:22.343795887] [20270] ERROR IPAModule ipa_module.cpp:279 ipa_vimc.so: IPA module is not an ELF file
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
The sequence number of captured frames is logged to the console with
padding to 6 characters to increase readability. The output is however
incorrect, as 123 is printed as 00012300000. This is caused by the
auto-space feature of QDebug, which inserts a space after every field.
This doesn't play well with stream format manipulation, as it ends up
padding the automatically inserted space the same way as the previous
argument.
This is a bug in Qt, work around it by formatting the sequence number
manually.
Fixes: 494da4467ddf ("qcam: Use Qt qInfo() and qWarning() logging facilities")
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
When starting the streaming, it might be that the call
to sensorInfo fails. This does not mean that the camera can't
stream so set 'ret' to 0.
Fixes: fd554f9dba31 ("libcamera: ipa: Add support for CameraSensorInfo")
Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
|
|
Improve the pre-push git hook script to reject commits without at least
one Reviewed-by line.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
|
|
Improve the pre-push git hook script to reject commits without the
committer's Signed-off-by line.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
|
|
When the IPA modules are installed, meson strips the DT_RPATH and
DT_RUNPATH from the binaries. This invalidates the signatures. Disable
installation of the .sign files and add an installation script to
regenerate them directly in the target directory. The .sign files still
need to be created at build time to support running IPA modules from the
build tree.
Two alternative approaches have been considered:
- meson could be taught a new target argument to preserve binary
compatibility by skipping any operation that modifies files. This has
been proposed in the #mesonbuild IRC channel. While this could be
interesting in the longer term, we need to fix the issue now.
- The module signatures could be computed on selected sections only.
While skipping the .dynamic section when signing may not cause
security issues, it would make signature generation and verification
more complex, and wasn't deemed worth it.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
|