When acquiring the media device, it is not necessary to match all entity names,
so remove it. Aditionally, we do not need to keep the MediaEntity pointers for
the Unicam and ISP devices stored within the PipelineHandlerRPi class. Instead
these can be stored locally in PipelineHandlerRPi::match().
PipelineHandlerRPi::registerCamera() now returns an int error code instead of a
boolean for pass/fail.
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Split out PipelineHandlerRPi::match() so that media device enumeration and
acquisition is separated from camera registration. The former logic remains
in PipelineHandlerRPi::match(), whereas the latter logic is moved into a new
PipelineHandlerRPi::registerCamera() member function.
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
For simplicity, the pipeline handler currently look at the maximum number of
buffers set in the StreamConfiguration by the user and allocate the same number
of internal buffers for all device nodes. This would likely overallocate buffers
for some nodes. Rework this logic to try and minimise overallcations without
compromising performance.
The key change is to mostly decouple the number of internal buffers allocated
from number of buffers requested by the user through the StreamConfiguration.
For ISP nodes, we only ever need 1 set of internal buffers, as the hardware runs
synchronous with the requests and IPA.
For Unicam nodes, allocate a minimum for 4 buffers (exported + internal), but
also require at least 2 internal buffers to minimise frame drops.
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The aspect ratio calculation divides two integer values then casts to a double.
This might reduce precision when scoring for aspect rato differences.
Fix this by casting the integer to a double before the division.
Reported-by: Coverity CID=361652
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The unicam driver no longer registers an embedded data node if the sensor does
not provide this stream. Account for this in the pipeline handler match routine
by not assuming it is always present.
Add a warning if Unicam and the CamHelper do not agree on the presense of sensor
embedded data, and disable its usage in these cases.
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Switch the pipeline handler to use the new Unicam media controller based driver.
With this change, we directly talk to the sensor device driver to set controls
and set/get formats in the pipeline handler.
This change requires the accompanying Raspberry Pi linux kernel change at
https://github.com/raspberrypi/linux/pull/4645. If this kernel change is not
present, the pipeline handler will fail to run with an error message informing
the user to update the kernel build.
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Jacopo Mondi <jacopo@jmondi.org>
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Sensor flips might change the Bayer order of the requested format. The existing
code would set a sensor format along with the appropriate Unicam and ISP input
formats, but reset the latter two on start() once the flips had been requested.
We can now set the sensor flips just before we set the sensor mode in
configure(), thereby not needing the second pair of format sets in start().
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
When the pipeline handler start() method is supplied with a NULL list
of controls, we send an empty control list to the IPA. When the IPA is
running in isolated mode the control list goes through the data
serializer, for which it must be marked correctly as a list of
"controls::controls", otherwise the IPA process will abort.
The IPA has a similar problem returning a control list in its
configure() method. We must be careful to initialise it properly even
when empty.
Signed-off-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The V4L2VideoDevice::toV4L2PixelFormat() function is incorrectly
implemented, as it will pick a multi-planar format if the device
supports the multi-planar API, even if only single-planar formats are
supported. This currently works because the implementation calls
V4L2PixelFormat::fromPixelFormat(), which ignores the multiplanar
argument and always returns a single-planar format.
Fixing this isn't trivial. As we don't need to support multi-planar V4L2
formats at this point, drop the function instead of pretending
everything is fine, and call V4L2PixelFormat::fromPixelFormat() directly
from pipeline handlers. As the single-planar case is the most common,
set the multiplanar argument to false by default to avoid long lines.
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>
Reviewed-by: Hirokazu Honda <hiroh@chromium.org>
As part of the effort to remove the CameraData class, migrate the
pipeline handler-specific camera data from CameraData to the
Camera::Private class.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Jacopo Mondi <jacopo@jmondi.org>
In order to allow subclassing Camera::Private in pipeline handlers, pass
the pointer to the private data to the Camera constructor, and to the
Camera::createCamera() function.
The Camera::Private id_ and streams_ members now need to be initialized
by the Camera constructor instead of the Camera::Private constructor, to
allow storage of the streams in a pipeline handler-specific subclass of
Camera::Private.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Jacopo Mondi <jacopo@jmondi.org>
When RPiCameraData::clearIncompleteRequests() clears out the request
queue during a stop condition, it unconditionally calls completeBuffer()
on all buffers in each request. This is wrong, as a buffer could have
already been completed as part of the current request, but the request
itself may not yet have completed.
Fix this by checking if the buffers in the request have been completed
before cancelling them.
Fixes: d372aaa10d ("pipeline: raspberrypi: Simplify RPiCameraData::clearIncompleteRequests()")
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Tested-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
With the addition of FrameBuffer::cancel(), the logic to clear and return
pending requests can be simplified by not having to queue all the request
buffers to the device before calling streamOff().
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Tested-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Jacopo Mondi <jacopo@jmondi.org>
Signed-off-by: Jacopo Mondi <jacopo@jmondi.org>
With the recent change to merge existing Request metadata with the
ControlList provided by the IPA in commit fcfb1dc02a ("libcamera:
raspberry: Report sensor timestamp"), we can now write the
controls::ScalerCrop value at the start of the pipeline instead of at
the end.
This change simplifies the logic slightly, and allows us to write all
metadata items to the Request in one place.
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: David Plowman <david.plowman@raspberrypi.com>
Tested-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Write the controls::SensorTimestamp value in the Request metadata when
the request is popped from the queue ready to run the pipeline. This
ensures that the timestamp is written to the correct Request item,
which may not be at the top of the queue when the Unicam buffer dequeue
occurs.
Fixes: fcfb1dc02a ("libcamera: raspberry: Report sensor timestamp")
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: David Plowman <david.plowman@raspberrypi.com>
Tested-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The controller algorithms currently run on every frame provided to the
IPA by the pipeline handler. This may be undesirable for very fast fps
operating modes where it could significantly increase the computation
cycles (per unit time) without providing any significant changes to the
IQ parameters. The added latencies could also cause dropped frames.
Pass the FrameBuffer timestamp to the IPA through the controls. This
timestamp will be used to rate-limit the controller algorithms to run
with a minimum inter-frame time given by a compile time constant,
currently set to 16.66ms. On startup, we don't rate-limit the algorithms
until after the number of frames required for convergence.
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Report the sensor's timestamp in the Request metadata by using the
Unicam::Image buffer timestamp as an initial approximation.
The buffer's timestamp is recorded at DMA-transfer time, and it does not
theoretically matches the 'start of exposure' definition, but when used
to compare two consecutive frames it gives an acceptable estimation of
the sensor frame period duration.
Reviewed-by: Hirokazu Honda <hiroh@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Naushir Patuck <naush@raspberrypi.com>
Signed-off-by: Jacopo Mondi <jacopo@jmondi.org>
When a Request is completed upon receiving the IPA produced metadata,
the metadata associated with the Request are over-written, deleting
the information set at output buffer completion, such as the
SensorTimestamp.
This commit applies to the RaspberryPi pipeline handler the same change
applied to IPU3 in commit 2ee3dd57e2fc ("libcamera: ipu3: Merge IPA
metadata controls").
Reviewed-by: Hirokazu Honda <hiroh@chromium.org>
Reviewed-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
Reviewed-by: Naushir Patuck <naush@raspberrypi.com>
Signed-off-by: Jacopo Mondi <jacopo@jmondi.org>
Rename ConfigInput to IPAConfig to be more consistent with the naming,
and remove ConfigInput::op, as it is never used.
Replace ConfigOutput with a ControlList type, as that is the only return
type from ipa::configure().
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Tested-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Separate out the in and out parameters in ipa::start() as they are not
the same. This function now takes in a ControlList and returns out a
struct StartConfig which holds a ControlList and drop frame count for
the pipeline handler to action.
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Tested-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Move the opening of the CamHelper from ipa::configure() to ipa::init().
This allows the pipeline handler to get the sensor specific parameters
in pipeline_handler::match() where the ipa is initialised.
Having the sensor parameters available earlier will allow selective
use of the embedded data node in a future change.
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Tested-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
If the ISP::Output0 stream has not been configured, we must enable it
with a default format and resolution for internal use. This is to allow
the pipeline handler data flow to be consistent, and allow the IPA to
receive statistics for the frame.
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
It is possible for the application to call pipeline_handler::configure()
multiple times, which would attempt to open the Unicam embedded data
node on every call. This would cause a warning message as the node
would have already been opened. Avoid this by tracking if the node
has previously been opened.
Note that this is a temporary fix since the open call for the Unicam
embedded data node will be moved from pipeline_handler::configure() to
pipeline_handler::match().
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
If an exposure time change adjusts the vblanking limits, and we set both
VBLANK and EXPOSURE controls through the VIDIOC_S_EXT_CTRLS ioctl, the
latter may fail if the value is outside of the limits calculated by the
old VBLANK value. This is a limitation in V4L2 and cannot be fixed by
setting VBLANK before EXPOSURE in a single VIDIOC_S_EXT_CTRLS ioctl.
The workaround here is to have the DelayedControls object mark the
VBLANK control as "priority write", which then write VBLANK separately
from (and ahead of) any other controls. This way, the sensor driver will
update the EXPOSURE control with new limits before the new values is
presented, and will thus be seen as valid.
To support this, a new struct DelayedControls::ControlParams is used in
the constructor to provide the control delay value as well as the
priority write flag.
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Tested-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
Tested-by: Jean-Michel Hautbois <jeanmichel.hautbois@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
[Kieran: Fix up trivial comments, merge conflicts]
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Now that we support returning int directly in addition to other output
parameters, improve the configure() function in the raspberrypi IPA
interface.
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The pipeline handler would enable and use the Unicam embedded data stream
even if the sensor did not support it. This was to allow a means to
pass exposure and gain values for the frame to the IPA in a synchronised
way.
The recent changes to get the pipeline handler to pass a ControlList
with exposure and gain values means this is no longer required. Disable
the use of the embedded data stream when a sensor does not support it.
This change also removes the mappedEmbeddedBuffers_ map as it is no
longer used.
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Tested-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
When running with sensors that had no embedded data, the pipeline handler
would fill a dummy embedded data buffer with gain/exposure values, and
pass this buffer to the IPA together with the bayer buffer. The IPA would
extract these values for use in the controller algorithms.
Rework this logic entirely by having a new RPiCameraData::BayerFrame
queue to replace the existing bayer queue. In addition to storing the
FrameBuffer pointer, this also stores all the controls tracked by
DelayedControls for that frame in a ControlList. This includes include
exposure and gain values. On signalling RPi::IPA_EVENT_SIGNAL_ISP_PREPARE
IPA event, the pipeline handler now passes this ControlList from the
RPiCameraData::BayerFrame queue.
The IPA now extracts the gain and exposure values from the ControlList
instead of using RPiController::MdParserRPi to parse the embedded data
buffer.
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Tested-by: David Plowman <david.plowman@raspberrypi.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Usage of std::move() on a const lvalue will produce a const rvalue,
which the compiler will bind to the const lvalue reference of the
constructor or assignement operator. This results in the object being
copied, not moved. Drop std::move() in that case as it's misleading.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Naushir Patuck <naush@raspberrypi.com>
Naushir Patuck
David Plowman
Laurent Pinchart
Kieran Bingham
Umang Jain
Hirokazu Honda
Jacopo Mondi
Sebastian Fricke
Paul Elder