For consistency with UniqueFD, rename the fd() function to get().
Renaming UniqueFD::get() to fd() would have been another option, but was
rejected to keep as close as possible to the std::shared_ptr<> and
std::unique_ptr<> APIs.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Jacopo Mondi <jacopo@jmondi.org>
Reviewed-by: Hirokazu Honda <hiroh@chromium.org>
Remove the verbose #ifndef/#define/#endif pattern for maintaining
header idempotency, and replace it with a simple #pragma once.
This simplifies the headers, and prevents redundant changes when
header files get moved.
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Jean-Michel Hautbois <jeanmichel.hautbois@ideasonboard.com>
"using namespace" in a header file propagates the namespace to
the files including the header file. So it should be avoided.
This removes "using namespace" in stream_options.h
Signed-off-by: Hirokazu Honda <hiroh@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The camera session keeps requeuing requests until the capture limit is
reached. This causes more request than the limit to complete, as there's
a queue of requests in flight. When capturing from multiple cameras
concurrently, this results in the captureDone signal being emitted for
every request completion after the limit is reached, instead of once per
camera session when reaching the limit.
Fix this by simply dropping any request that completes after the limit
is reached. We could instead avoid requeuing more requests than needed
to reach the limit, but that may cause request starvation in pipelines,
which are currently not handled consistently (or correctly).
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The FileSink class constructs stream names internally the same way that
the CameraSession does, except that it fails to add the camera name.
This results in files being written without the camera name.
This could be fixed in FileSink, but we would still duplicate code to
construct stream names. Pass the stream names map from CameraSession to
FileSink instead, and store it internally.
Fixes: 02001fecb0 ("cam: Turn BufferWriter into a FrameSink")
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
When creating a DRM frame buffer, the dmabufs for the planes are
imported as GEM objects. For multi-planar formats, all planes may use
the same dmabuf, which results in multiple imports. This doesn't cause
any issue at import time, as DRM detects this situation and returns the
same GEM object. However, when destroying the frame buffer, the same GEM
object ends up being closed multiple times, which generates an error.
Fix this by avoiding multiple imports of the same dmabuf for the same
frame buffer. While the issue may theoretically occur with identical
dmabufs for different frame buffers, this is quite unlikely and is thus
not addressed.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Jean-Michel Hautbois <jeanmichel.hautbois@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The stride is not always identical for all planes for multi-planar
formats. Semi-planar YUV formats without horizontal subsampling often
have a chroma stride equal to twice the luma stride, and tri-planar YUV
formats with a 1/2 horizontal subsampling often have a chroma stride
equal to half the luma stride. This isn't correctly taken into account
when creating a DRM frame buffer, as the same stride is set for all
planes.
libcamera doesn't report per-plane stride values yet, but uses chroma
strides that match the above description for all currently supported
platforms. Calculation the chrome strides appropriately in the KMSSink
class, and pass them to DRM::createFrameBuffer().
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Jean-Michel Hautbois <jeanmichel.hautbois@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The new Image class represents a multi-planar image with direct access
to pixel data. It currently duplicates the function of the
MappedFrameBuffer class which is internal to libcamera, and will serve
as a design playground to improve the API until it is considered ready
to be made part of the libcamera public API.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Hirokazu Honda <hiroh@chromium.org>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The number of metadata planes should always match the number of frame
buffer planes. Enforce this by making the vector private and providing
accessor functions.
As this changes the public API, update all in-tree users accordingly.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Jean-Michel Hautbois <jeanmichel.hautbois@ideasonboard.com>
Reviewed-by: Hirokazu Honda <hiroh@chromium.org>
To allow testing capture from multiple cameras concurrently, turn the
--camera option into an array option, and create one CameraSession per
specified camera. The code is adapted to iterate over the sessions
vector instead of handling a single session. Thanks to all the
refactoring previously performed, changes are minimal.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
To prepare for multi-camera support, extend the file naming scheme for
captured frames to include the camera index in addition to the stream
name and frame number.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Use the new hierarchical options feature of the option parser to turn
camera-related option (--capture, --file, --stream, --strict-formats and
--metadata) into children of the --camera option. As an added bonus, we
don't need to check anymore if a camera has been specified when capture
is requested, as that's now enforced by the option parser.
This change prepares for support of multiple cameras.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The value of the --file option is the full name of the file to which
captured frames are written. To write files to a specific directory with
the default naming scheme, the "frame-#.bin" name has to appear at the
end of the file name. Simplify usage of the option by allowing --file to
specify a directory only. If the file name ends with a '/', the default
"frame-#.bin" file name is automatically appended.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reorganize the run() function to make it more readable, and merge the
two event loops into one as capture and hotplug monitoring don't have to
be mutually exclusive.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The three CamApp functions listControls(), listProperties() and
infoConfiguration() operate on a camera. They would thus be better
placed in the CameraSession class. Move them there. As they now have no
error to return anymore, make them void functions.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Move creation of the camera session and setup of the hotplug monitoring
from the init() function to the run() function, to group all the code
that performs operations based on command line options in a single
place. The cleanup() call on session creation failure isn't required
anymore, as the cleanup() function is called unconditionally upon return
from run().
This change allows merging two different code blocks related to hotplug
monitoring.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The CamApp::cameraName() doesn't need to access any member of the CamApp
class. Make it static. While at it, drop an unneeded const from the
return value.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The EventLoop::exec() function returns the exit code of the event loop,
not an error status. Drop the corresponding error check and error
message.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Continue moving towards making the CameraSession class the central point
to handle a camera by moving the camera acquire operation. A new
CameraSession::camera() function is needed to allow access to the camera
from CamApp.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Move the local CameraSession variable from the CamApp::run() function to
a member variable of the CamApp class, created in CamApp::init(). This
is a first step towards moving code to the CameraSession class.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Store the BufferWriter and FrameBufferAllocator pointers in
std::unique_ptr<> instances to simplify memory management and avoid
leaks.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
To prepare for multiple concurrent camera sessions, move the event loop
exec() call from the CameraSession class to the CamApp class.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Make exiting the event loop the responsibility of the application, not
the camera session, to prepare for support of multiple camera sessions.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Don't pass the event loop to the CameraSession constructor, as passing
this global object explicitly isn't a design that can scale.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Rename the Capture class to CameraSession, to prepare for multi-camera
support that will gather more camera-related operations than capture in
that class.
While at it, remove an unneeded blank line.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The options parser currently ignores non-option arguments silently,
which is confusing. It's a common error to run 'cam -c1 -C 10' and
expect only 10 frames to be captured. As the -C option takes an optional
argument, the number 10 is interpreted as a non-option argument instead
of the argument to the -C option.
Fail parsing with an appropriate message and print usage information
when a non-option argument is found. The parser may be extended later to
accept non-option arguments when the application has a use for them.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Jacopo Mondi <jacopo@jmondi.org>
The OptionValue toKeyValues() and toArray() functions return a copy of
the values. This is unnecessary, and can cause use-after-free issues
when taking references to the return values. Return references instead
to optimize the implementation and avoid issues.
The behaviour of the two functions is now undefined in case of an option
type mismatch. The current implementation catches this with an
assertion.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Jacopo Mondi <jacopo@jmondi.org>
The roles() and updateConfiguration() functions check if the OptStream
OptionValue they receive is empty by first casting it to an array. To
prepare for the toArray() function not allowing such a cast when the
option value is empty, test if the option value is empty instead.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Jacopo Mondi <jacopo@jmondi.org>
Add a convenience helper to check if an option value is empty, based on
the value type. This is useful as a shortcut syntax to check if an
option has been set.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Jacopo Mondi <jacopo@jmondi.org>
While OptionValue cast operators to int and std::string allow useful
shortcut syntaxes, the cast operators to KeyValueParser::Options and
std::vector<OptionValue> are less useful. A an explicit static_cast call
would be more cumbersome to write than an explicit .toKeyValues() or
toArray(), and implicit cast hide too much of what's going on.
Drop those two cast operators, and replace the only implicit cast
occurrence with .toKeyValues(). While at it, drop the local opts
variable in StreamKeyValueParser::roles() as it isn't used.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
