Just return an `std::unique_ptr` constructed from an empty
initializer instead of doing a `reset()` on the existing
`config` variable and returning that. This is simpler.
Signed-off-by: Barnabás Pőcze <pobrn@protonmail.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Objects of type `Object` and `Thread` have address identities, so they
should not be just moved/copied. And the special member functions
generated by the compiler do not do the right thing. So delete them.
Signed-off-by: Barnabás Pőcze <pobrn@protonmail.com>
Reviewed-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The mentioned commit mistakenly introduced a new variable for storing
the camera instead of just assigining to the global variable defined
earlier in the tutorial. Fix that by making it an assignment.
Bug: https://bugs.libcamera.org/show_bug.cgi?id=252
Fixes: e77a275110 ("treewide: Query list of cameras just once")
Signed-off-by: Barnabás Pőcze <pobrn@protonmail.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
inputBufferReady ready signal in the simple pipeline is handled in the
pipeline handler thread. outputBufferReady and ispStatsReady signals
should be handled there too.
Rather than relying on the user of the SoftwareIsp instance, let
SoftwareIsp inherits Object. SoftwareIsp serves as a signal proxy, the
signals above are emitted from signal handlers. This means that if
SoftwareIsp inherits Object then the slots are invoked in SoftwareIsp
thread. Which is the camera manager thread because the SoftwareIsp
instance is created there.
Signed-off-by: Milan Zamazal <mzamazal@redhat.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>
EINVAL should be returned only when the requested format is unsupported.
On errors when writing the data, let's return EIO, which is the closest
description of the situation when we don't inspect it more.
Signed-off-by: Milan Zamazal <mzamazal@redhat.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
<errno.h> should be included in the ppm writer, as the source of the
error code constants used there.
Signed-off-by: Milan Zamazal <mzamazal@redhat.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The check for the number of outputs is done in a loop over the outputs.
It should be moved out of the loop as it's not loop specific and is just
repeated there.
This is a cosmetic change not changing any functionality.
Signed-off-by: Milan Zamazal <mzamazal@redhat.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The rkisp1 and rpi pipeline handlers duplicate code to handle the
LIBCAMERA_RKISP1_TUNING_FILE and LIBCAMERA_RPI_TUNING_FILE environment
variables that override tuning file selection. Move the common code to
IPAProxy::configurationFile() to avoid the duplication, and make the
feature available to all pipeline handlers with the same behaviour.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
Indentation in a doxygen comment is wrong, fix it.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
Even though the new AeEnable control internally switches on and off the
sub-controls (ExposureTimeMode and AnalogueGainMode), it still needs to
be declared as available. Report this control as available in the
rpi IPA.
Support for the control does not need to be added as it is handled by
the Camera class. It does not need to be handled in metadata either as
the new version of AeEnable is not returned in metadata.
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
Reviewed-by: Naushir Patuck <naush@raspberrypi.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Even though the new AeEnable control internally switches on and off the
sub-controls (ExposureTimeMode and AnalogueGainMode), it still needs to
be declared as available. Report this control as available in the
rkisp1 IPA.
Support for the control does not need to be added as it is handled by
the Camera class. It does not need to be handled in metadata either as
the new version of AeEnable is not returned in metadata.
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Handle the AeEnable under the hood in the Camera class, such that
AeEnable activates ExposureTimeMode and AnalogueGain together. This
allows applications the convenience of setting auto/manual mode of all
of the AE-related controls, as well as protecting applications against a
nasty behavior change if an aperture control is added in the future.
This also moves common handling code out of the IPA.
While we also want to inject AeEnable in Camera::controls() so that IPAs
don't have to report it, it is technically difficult at the moment as
ControlInfoMaps are not easily modifiable.
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
In the redesign of the AE-related controls, the AeEnable control was
intended to be removed in favor of more specific sub-controls for
analogue gain and exposure time. However this will cause problems if
aperture sub-controls are introduced, and an application from a
pre-aperture era uses a camera that supports aperture.
If there is no AeEnable control, then a pre-aperture era application
might set analogue gain and exposure time to manual, while aperture
silently stays auto since that's the default mode. Thus aperture would
be uncontrollable by the application.
With an AeEnable control, then a pre-aperture era application can set
AeEnable to manual, and under the hood all three of analogue gain and
exposure time and aperture will be set to manual. The application won't
be able to set the manual aperture, however.
Although the above scenario is expected to be rare, the scenario with an
AeEnable control seems less detrimental. With an AeEnable control at
least the aperture would be static at a reasonably usable value, whereas
without an AeEnable the aperture would be more-or-less uncontrolable and
could go to extreme values as the AEGC algorithm tries to compensate for
the manual analogue gain and exposure time values.
Thus we redefine the AeEnable control, available only as a control and
not in metadata. It will be preprocessed by the Camera class so that the
relevant sub-controls are set. No pipline handler nor IPA shall act on
the AeEnable control. The IPA still has to report the control as
available, however.
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Now that the codebase has been ported to use the new AEGC controls
remove the definition of AeLocked. AeEnable is not be removed as it will
be redefined.
Signed-off-by: Jacopo Mondi <jacopo@jmondi.org>
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Since AeEnable will be replaced with ExposureTimeMode and
AnalogueGainMode so that the two can be set between auto/manual
independently, update the gstreamer control ids generation to conform
with this.
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Nicolas Nicolas <nicolas.dufresne@collabora.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The newly introduced controls to drive the AEGC algorithm allow
controlling the computation of the exposure time and analogue gain
separately.
Augument the RkISP1 AEGC implementation to handle the exposure and gain
controls separately using the new controls.
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The newly introduced controls to drive the AEGC algorithm allow to
control the computation of the exposure time and analogue gain
separately.
The RPi AEGC implementation already computes the shutter and gain values
separately but does not expose separate functions to control them.
Augment the AgcAlgorithm interface to allow pausing/resuming the shutter
and gain automatic computations separately and plumb them to the newly
introduced controls.
Add safety checks to ignore ExposureTime and AnalogueGain values if the
algorithms are not paused, and report the correct AeState value by
checking if both algorithms have been paused or if they have converged.
Signed-off-by: Jacopo Mondi <jacopo@jmondi.org>
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
Reviewed-by: Naushir Patuck <naush@raspberrypi.com>
Acked-by: David Plowman <david.plowman@raspberrypi.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Replace the deprecated AeEnable control with DebugMetadataEnable
in ipa_data_serialization test. We use DebugMetadataEnable instead of
one of the controls replacing AeEnable as they are not boolean controls.
Signed-off-by: Jacopo Mondi <jacopo@jmondi.org>
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Port the UVC pipeline handler to use the new ExposureTimeMode control
when processing Camera controls in place of the AeEnable control.
The V4L2_CID_EXPOSURE_AUTO control allows 4 possible values, which
map to ExposureTimeModeAuto and ExposureTimeModeManual.
Signed-off-by: Jacopo Mondi <jacopo@jmondi.org>
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Document the design and rationale for the AE-related controls.
Also add documentation for the controls.
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Signed-off-by: Jacopo Mondi <jacopo@jmondi.org>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Introduce the AeState, ExposureTimeMode and AnalogueGainMode controls
to model the AEGC algorithm block.
The three controls allow applications to select the exposure time and
analogue gain computation calculation mode (auto vs manual)
independently from one another, while the AeState control reports the
global state for the AEGC algorithm.
The new controls are meant to replace the existing AeEnable and AeLocked
controls, which are momentarily kept not to break compilation of
platforms making use of them.
Bug: https://bugs.libcamera.org/show_bug.cgi?id=42
Bug: https://bugs.libcamera.org/show_bug.cgi?id=43
Bug: https://bugs.libcamera.org/show_bug.cgi?id=47
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
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>
gcc 13.3.0, as provided by buildroot 2024.11.1, chokes when compiling
v4l2_subdevice.cpp with ASan enabled, due to the usage of the C++
standard library regex header:
In file included from /host/aarch64-buildroot-linux-gnu/include/c++/13.3.0/functional:59,
from ../../include/libcamera/base/utils.h:12,
from ../../include/libcamera/base/log.h:15,
from ../../include/libcamera/internal/v4l2_subdevice.h:20,
from ../../src/libcamera/v4l2_subdevice.cpp:8:
In constructor ‘std::function<_Res(_ArgTypes ...)>::function(std::function<_Res(_ArgTypes ...)>&&) [with _Res = bool; _ArgTypes = {char}]’,
inlined from ‘std::__detail::_State<_Char_type>::_State(std::__detail::_State<_Char_type>&&) [with _Char_type = char]’ at /host/aarch64-buildroot-linux-gnu/include/c++/13.3.0/bits/regex_automaton.h:149:4,
inlined from ‘std::__detail::_StateIdT std::__detail::_NFA<_TraitsT>::_M_insert_subexpr_begin() [with _TraitsT = std::__cxx11::regex_traits<char>]’ at /host/aarch64-buildroot-linux-gnu/include/c++/13.3.0/bits/regex_automaton.h:281:24:
/host/aarch64-buildroot-linux-gnu/include/c++/13.3.0/bits/std_function.h:405:42: error: ‘*(std::function<bool(char)>*)((char*)&__tmp + offsetof(std::__detail::_StateT, std::__detail::_State<char>::<unnamed>.std::__detail::_State_base::<unnamed>)).std::function<bool(char)>::_M_invoker’ may be used uninitialized [-Werror=maybe-uninitialized]
405 | : _Function_base(), _M_invoker(__x._M_invoker)
| ~~~~^~~~~~~~~~
In file included from /host/aarch64-buildroot-linux-gnu/include/c++/13.3.0/regex:65,
from ../../src/libcamera/v4l2_subdevice.cpp:11:
/host/aarch64-buildroot-linux-gnu/include/c++/13.3.0/bits/regex_automaton.h: In member function ‘std::__detail::_StateIdT std::__detail::_NFA<_TraitsT>::_M_insert_subexpr_begin() [with _TraitsT = std::__cxx11::regex_traits<char>]’:
/host/aarch64-buildroot-linux-gnu/include/c++/13.3.0/bits/regex_automaton.h:279:17: note: ‘__tmp’ declared here
279 | _StateT __tmp(_S_opcode_subexpr_begin);
| ^~~~~
In member function ‘bool std::_Function_base::_M_empty() const’,
inlined from ‘std::function<_Res(_ArgTypes ...)>::operator bool() const [with _Res = bool; _ArgTypes = {char}]’ at /host/aarch64-buildroot-linux-gnu/include/c++/13.3.0/bits/std_function.h:574:25,
inlined from ‘std::function<_Res(_ArgTypes ...)>::function(std::function<_Res(_ArgTypes ...)>&&) [with _Res = bool; _ArgTypes = {char}]’ at /host/aarch64-buildroot-linux-gnu/include/c++/13.3.0/bits/std_function.h:407:6,
inlined from ‘std::__detail::_State<_Char_type>::_State(std::__detail::_State<_Char_type>&&) [with _Char_type = char]’ at /host/aarch64-buildroot-linux-gnu/include/c++/13.3.0/bits/regex_automaton.h:149:4,
inlined from ‘std::__detail::_StateIdT std::__detail::_NFA<_TraitsT>::_M_insert_subexpr_begin() [with _TraitsT = std::__cxx11::regex_traits<char>]’ at /host/aarch64-buildroot-linux-gnu/include/c++/13.3.0/bits/regex_automaton.h:281:24:
/host/aarch64-buildroot-linux-gnu/include/c++/13.3.0/bits/std_function.h:247:37: error: ‘*(const std::_Function_base*)((char*)&__tmp + offsetof(std::__detail::_StateT, std::__detail::_State<char>::<unnamed>.std::__detail::_State_base::<unnamed>)).std::_Function_base::_M_manager’ may be used uninitialized [-Werror=maybe-uninitialized]
247 | bool _M_empty() const { return !_M_manager; }
| ^~~~~~~~~~
/host/aarch64-buildroot-linux-gnu/include/c++/13.3.0/bits/regex_automaton.h: In member function ‘std::__detail::_StateIdT std::__detail::_NFA<_TraitsT>::_M_insert_subexpr_begin() [with _TraitsT = std::__cxx11:
/host/aarch64-buildroot-linux-gnu/include/c++/13.3.0/bits/regex_automaton.h:279:17: note: ‘__tmp’ declared here
279 | _StateT __tmp(_S_opcode_subexpr_begin);
| ^~~~~
This is a false positive that previously occurred with gcc 12.1.0 and
was fixed in 12.2 (see https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105562),
and it seems to have now reappeared. The bug report indicates that the
probability of such false positives increase when using sanitizers. As
this isn't caught by CI, which compiles libcamera with gcc 13.3.0, the
chance that such compilation failures will appear in environments
without a clear pattern is relatively high. Work around the problem by
disabling the warning around the inclusiong of the regex header.
If the regex header needs to be included in other source files, creating
a wrapped in libcamera-base may be a cleaner alternative.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
As DmaSyncer disables the copy c'tor, the move c'tor will be used
instead. This leaves some DmaSyncers with invalid SharedFDs. They should
avoid syncing with invalid file descriptors in the d'tor.
Fixes: 545046a41e ("DmaBufAllocator: Make DmaSyncer non-copyable")
Signed-off-by: Harvey Yang <chenghaoyang@chromium.org>
Tested-by: Milan Zamazal <mzamazal@redhat.com>
Reviewed-by: Milan Zamazal <mzamazal@redhat.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Install the example configuration file of the virtual pipeline
handler as it serves documentation purposes, and to make the
virtual pipeline handler easily usable in CI.
Nonetheless, the file is installed with the ".example" suffix
so that it will not be used by default, to avoid cluttering
the camera lists of users whose distributions decide to
enable the virtual pipeline handler.
The file is installed in the proper location for convenience:
(1) is is easier to use it in the CI;
(2) users need not browse documentation to determine where
they should place the file.
Signed-off-by: Barnabás Pőcze <pobrn@protonmail.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Harvey Yang <chenghaoyang@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The virtual pipeline handler prints an error message when its
configuration file can't be opened. Not providing a configuration file
is the default method to disable virtual cameras, so this error is
confusing for users. Replace it with a debug message when the
configuration file is not found, and keep an error message when it
exists but can't be opened.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Tested-by: Julien Vuillaumier <julien.vuillaumier@nxp.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The PipelineHandler::configurationFile() function prints an error
message when no configuration file is found. It can be useful for
pipeline handlers to silence the lookup operation and handle errors
themselves. Add a silent parameter to the function to enable this.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Tested-by: Julien Vuillaumier <julien.vuillaumier@nxp.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Consider `pipelines=auto,virtual`. Previously that would select
everything that `auto` would, but not actually enable the `virtual`
pipeline handler because the `pipelines` list was reset.
Bug: https://bugs.libcamera.org/show_bug.cgi?id=247
Signed-off-by: Barnabás Pőcze <pobrn@protonmail.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The v4l2 compatibility layer does not have external dependencies,
the usual benefits of a feature option do not apply. The main
motivation behind this change is making use of the `auto_features`
meson option that can change all feature options set to "auto"
by default to the desired value.
This can be useful for two reasons: (1) using auto_features=disabled
and then building up the list of features that are desired in the
particular build; (2) using auto_features=enabled to achieve
(usually) more testing and compilation coverage.
Signed-off-by: Barnabás Pőcze <pobrn@protonmail.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
If `FrameBufferAllocator::allocate()` causes the construction to be
aborted, the allocated `GstLibcameraAllocator` will not be
deallocated properly. Use `g_autoptr()` to address this.
`g_steal_pointer()` could only be used in glib 2.68 or later because
earlier it evaluates to a pointer-to-void in C++, which would necessitate
a `static_cast`.
Signed-off-by: Barnabás Pőcze <pobrn@protonmail.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
`FrameBufferAllocator::allocate()` might return a negative error code,
but currently this is handled the same way as success. So instead
of continuing, abort the construction of the allocator object.
Signed-off-by: Barnabás Pőcze <pobrn@protonmail.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Previously we call Thread::setThreadAffinityInternal in
Thread::startThread. The purpose was to avoid the main workload being
run on incorrect CPUs. This leads to a race condition of setting
`Thread::thread_` in `Thread::start()` and accessing
`Thread::setThreadAffinityInternal` though.
This patch moves the call after the construction of std::thread to avoid
the race condition. The downside is that the first tasks, if any, upon
starting a thread might be run on incorrect CPUs.
Fixes: 4d9db06d66 ("libcamera: add method to set thread affinity")
Signed-off-by: Harvey Yang <chenghaoyang@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The error message for missing direction field prints the direction value
(usually 'None') instead of the name of the field 'direction'. Fix this.
Fixes: 39fe4ad968 ("utils: codegen: controls.py: Parse direction information")
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Many libcamera headers that use standard C integer types do not include
stdint.h. Fix the omission.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Without the change the build fails on upcoming `gcc-15` as:
In file included from ../src/libcamera/dma_buf_allocator.cpp:9:
../include/libcamera/internal/dma_buf_allocator.h:66:19: error: 'uint64_t' has not been declared
66 | void sync(uint64_t step);
| ^~~~~~~~
Signed-off-by: Sergei Trofimovich <slyich@gmail.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
__nodiscard was introduced for compatibility with C++14.
In C++17, there is an official attribute: [[nodiscard]].
Moreover, some libc implementations (like bionic) already define the
__nodiscard macro [1].
Since:
- libcamera builds with cpp_std=c++17
- [[nodiscard]] is already used in the android HAL (exif)
We should replace all usage __nodiscard of by [[nodiscard]] for
consistency.
Do the replacement and remove the no longer used compiler.h.
[1] 3254860
Signed-off-by: Mattijs Korpershoek <mkorpershoek@baylibre.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>
Some sensors support producing and transmitting embedded data over a
stream separate from the image stream. Add support for this feature in
the CameraSensor interface, and implement it for the CameraSensorRaw
class. The CameraSensorLegacy uses the default stub implementation, as
the corresponding kernel drivers don't support embedded data.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Add a new CameraSensorRaw implementation of the CameraSensor interface
tailored to devices that implement the new V4L2 raw camera sensors API.
This new class duplicates code from the CameraSensorLegacy class. The
two classes will be refactored to share code.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
With support for metadata in the streams API, the v4l2_meta_format
structure has been extended with width, height and bytesperline fields.
Support them in the V4L2VideoDevice getFormat() and setFormat()
functions is the video device is meta capture device and if the
pixel format is one of the generic line-based metadata formats.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Signed-off-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Support the newly introduced V4L2 media bus formats for metadata. This
includes generic metadata formats, and two sensor-specific embedded data
formats.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Signed-off-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Add the definition for the generic line based pixelformats.
The formats has been added in upstream Linux by commit
1d9215233958 ("media: uapi: v4l: Add generic 8-bit metadata format
definitions") which got merged in Linux v6.10.
The formats however are not yet available to userspace, as they
have been made only available to the kernel by commit
d69c8429ea80 ("media: uapi: v4l: Don't expose generic metadata formats
to userspace") to let the line-based metadata support stabilize before
allowing applications to use it.
With the forthcoming completion of the line-based metadata upstreaming
manually add the generic line based pixel format to prepare libcamera
to support them.
Signed-off-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Acked-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Update the kernel headers with the definition of two device-specific
line-based metadata formats, and with the definition of the
MEDIA_PAD_FL_INTERNAL and V4L2_SUBDEV_ROUTE_FL_IMMUTABLE flags.
The new definitions will allow to support handling line-based
metadata streams exposed by the sensor driver through an
internal sink pad.
While the changes have not yet been collected in the official
linux-media tree, they're available in the 'metadata' branch of
https://git.linuxtv.org/sailus/media_tree.git, at revision:
f9bbbd9a696d ("media: Documentation: Add binning and sub-sampling
controls")
Signed-off-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Acked-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Add a CamHelper::setHwConfig() helper used by the IPA to set the
hardware configuration in use by the pipeline. This will be needed by
the IMX500 camera helper in a future commit to determine if the
metadata buffer is strided.
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: David Plowman <david.plowman@raspberrypi.com>
Acked-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
This property (dataBufferStrided) indicates if the CSI-2 hardware writes
to the embedded/metadata buffer directly, or if it treats the buffer
like an image buffer and strides the metadata lines.
Unicam writes this buffer strided, while the PiSP Frontend writes to it
directly. This information will be relevant to data parsers in the
helpers where the data is structured in lines.
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: Kieran Bingham <kieran.bingham@ideasonboard.com>
Use an r-value reference in set() and setLocked(), allowing more
efficient metadata handling with std::forward and std::move if needed.
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: Kieran Bingham <kieran.bingham@ideasonboard.com>
As another Starvis sensor, it is near identical to imx290/327.
Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: Naushir Patuck <naush@raspberrypi.com>
Reviewed-by: David Plowman <david.plowman@raspberrypi.com>
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
As DmaSyncer does sync start/end in the c'tor/d'tor, copying a DmaSyncer
instance would trigger sync end earlier than expected. This patch makes
it non-copyable to avoid the issue.
Fixes: 39482d59fe ("DmaBufAllocator: Add Dma Buffer synchronization function & helper class")
Signed-off-by: Harvey Yang <chenghaoyang@chromium.org>
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>
After the initial generation, when a frame is requested,
the test pattern generator rotates the image left by 1 column.
The current approach has two shortcomings:
(1) it allocates a temporary buffer to hold one column;
(2) it swaps two columns at a time.
The test patterns are simple ARGB images, in row-major order,
so doing (2) works against memory prefetching. This can be
addressed by doing the rotation one row at a time as that way
the image is addressed in a purely linear fashion. Doing so also
eliminates the need for a dynamically allocated temporary buffer,
as the required buffer now only needs to hold one sample,
which is 4 bytes in this case.
In an optimized build, this results in about a 2x increase in the
number of frames per second as reported by `cam`. In an unoptimized,
ASAN and UBSAN intrumented build, the difference is even bigger,
which is useful for running lc-compliance in CI in a reasonable time.
Signed-off-by: Barnabás Pőcze <pobrn@protonmail.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Tested-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
