[libcamera-devel] [PATCH 4/4] libcamera: pipeline: Add Intel IPU3 pipeline
Kieran Bingham
kieran.bingham at ideasonboard.com
Tue Jan 15 20:35:19 CET 2019
Hi Jacopo,
On 15/01/2019 17:37, Jacopo Mondi wrote:
> Hi Kieran,
>
> On Tue, Jan 15, 2019 at 03:15:23PM +0000, Kieran Bingham wrote:
>> Hi Jacopo,
>>
>> On 15/01/2019 14:07, Jacopo Mondi wrote:
>>> Add a pipeline handler for the Intel IPU3 device.
>>>
>>> The pipeline handler creates a Camera for each image sensor it finds to be
>>> connected to an IPU3 CSI-2 receiver, and enables the link between the two.
>>>
>>> Tested on Soraka, listing detected cameras on the system, verifying the
>>> pipeline handler gets matched and links properly enabled.
>>>
>>
>> This looks like a really good start, and I can start to understand how
>> the matching is going on now.
>>
>> Aside from some discussion below I don't think there's anything really
>> major to block this IMO.
>>
>> I left in my text about putting ->release in the destructor in case it
>> was actually relevant - but I think reading further down it became clear
>> that what I was suggesting is not a viable structure to the code.
>>
>> Reviewed-by: Kieran Bingham <kieran.bingham at ideasonboard.com>
>>
>>> Signed-off-by: Jacopo Mondi <jacopo at jmondi.org>
>>> ---
>>> src/libcamera/pipeline/ipu3/ipu3.cpp | 249 ++++++++++++++++++++++++
>>> src/libcamera/pipeline/ipu3/meson.build | 3 +
>>> src/libcamera/pipeline/meson.build | 2 +
>>> 3 files changed, 254 insertions(+)
>>> create mode 100644 src/libcamera/pipeline/ipu3/ipu3.cpp
>>> create mode 100644 src/libcamera/pipeline/ipu3/meson.build
>>>
>>> diff --git a/src/libcamera/pipeline/ipu3/ipu3.cpp b/src/libcamera/pipeline/ipu3/ipu3.cpp
>>> new file mode 100644
>>> index 0000000..4c24c79
>>> --- /dev/null
>>> +++ b/src/libcamera/pipeline/ipu3/ipu3.cpp
>>> @@ -0,0 +1,249 @@
>>> +/* SPDX-License-Identifier: LGPL-2.1-or-later */
>>> +/*
>>> + * Copyright (C) 2019, Google Inc.
>>> + *
>>> + * ipu3.cpp - Pipeline handler for Intel IPU3
>>> + */
>>> +
>>> +#include <map>
>>> +#include <vector>
>>> +
>>> +#include <libcamera/camera.h>
>>> +
>>> +#include "device_enumerator.h"
>>> +#include "log.h"
>>> +#include "media_device.h"
>>> +#include "pipeline_handler.h"
>>> +
>>> +namespace libcamera {
>>> +
>>> +class PipelineHandlerIPU3 : public PipelineHandler
>>> +{
>>> +public:
>>> + PipelineHandlerIPU3();
>>> + ~PipelineHandlerIPU3();
>>> +
>>> + bool match(DeviceEnumerator *enumerator);
>>> +
>>> + unsigned int count();
>>> + Camera *camera(unsigned int id) final;
>>> +
>>> +private:
>>> + MediaDevice *cio2_;
>>> + MediaDevice *imgu_;
>>> +
>>> + unsigned int numCameras_;
>>> + std::map<unsigned int, Camera *> cameras_;
>>> +
>>> + unsigned int registerCameras();
>>> +};
>>> +
>>> +PipelineHandlerIPU3::PipelineHandlerIPU3()
>>> + : cio2_(nullptr), imgu_(nullptr), numCameras_(0)
>>> +{
>>> +}
>>> +
>>> +PipelineHandlerIPU3::~PipelineHandlerIPU3()
>>> +{
>>> + if (cio2_)
>>> + cio2_->release();
>>
>> Does the MediaDevice destructor not do this call to ->release()?
>> (And should it?)
>>
>> <edit - I read below... :) >
>>
>> Then these two would just be:
>>
>> delete cio2_;
>> delete imgu_;
>>
>> BTW: I'm not objecting to a direct call to release() here - just
>> considering options.
>
> As you found out below, I shall not delete MediaDevices from here.
>
>>
>>
>>> +
>>> + if (imgu_)
>>> + imgu_->release();
>>> +
>>> + for (auto camera : cameras_) {
>>
>> Reading here - it's not clear what a camera.second is?
>> I'll go dig in the code - but as it's not obvious (unless I'm missing
>> something obvious) - perhaps it needs some description ?
>>
>
> Cameras_' a std::map, just look at the type of the template.
> I can add a small comments, even if I have commented extensively on
> the indexing of cameras in the class..
So I'm still really confused by the naming.
Why is it called a second? Does it represent a period of time? Is it a
second instance of the camera? Why does the camera need a second instance...
OH. *it's not a libcamera::Camera*
You know what - I think I'm afraid I've just validated Laurent's
reasoning not to use auto.
I read: for (auto camera : cameras_) { ... as
'for each "libcamera::Camera" in cameras_ ... so I had no idea what a
'second' was.
But we're not dealing with libcamera::Camera here - we're dealing with a
std::pair from the map iterator.
I fear that's really not clear without naming the type here...
>>
>>> + if (!camera.second)
>>> + continue;
>>> +
>>> + camera.second->put();
>>> +
>>> + /*
>>> + * FIXME
>>> + * The lifetime management of Camera instances will be
>>> + * soon changed: as of now, the handler creates cameras, and
>>> + * -shall- destroy them as well to avoid leaks.
>>> + */
>>> + delete camera.second;
>>
>> Although - perhaps this '.second' is going to disappear - so it doesn't
>> matter.
>>
>
> Lifetime management of cameras is not IPU3 specific. The use of a map
> to index cameras according to their CSI-2 receiver indexes (so that
> camera connected to CSI-2 0 always comes first than camera connected
> to CSI-2 3) is specific of this implementation. If that does not get
> changed after this review round, the '.second' stays.
>
> I could:
> Camera *cam = camera->second;
Yes - that would make things a lot clearer too.
Perhaps we could also rename the iterator?
>
>>> + }
>>> +
>>> + cio2_ = nullptr;
>>> + imgu_ = nullptr;
>>> + cameras_.clear();
>>> +}
>>> +
>>> +unsigned int PipelineHandlerIPU3::count()
>>> +{
>>> + return numCameras_;
>>> +}
>>> +
>>> +Camera *PipelineHandlerIPU3::camera(unsigned int id)
>>> +{
>>> + if (id >= numCameras_)
>>> + return nullptr;
>>> +
>>> + /*
>>> + * The requested camera id does not match the key index used to store
>>> + * Camera instances in the 'cameras_' map.
>>> + *
>>> + * The 'id' argument represent the n-th valid cameras registered
>>> + * in the system, while the indexing key is the CSI-2 receiver index
>>> + * the camera sensor is associated to, and some receiver might have no
>>> + * camera sensor connected.
>>> + */
>>> + for (auto it = cameras_.begin(); it != cameras_.end(); ++it, --id) {
>>> + if (id == 0)
>>> + return (*it).second;
>>> + }
>>> +
>>> + return nullptr;
>>> +}
>>> +
>>> +bool PipelineHandlerIPU3::match(DeviceEnumerator *enumerator)
>>> +{
>>> + DeviceMatch cio2_dm("ipu3-cio2");
>>> + cio2_dm.add("ipu3-csi2 0");
>>> + cio2_dm.add("ipu3-cio2 0");
>>> + cio2_dm.add("ipu3-csi2 1");
>>> + cio2_dm.add("ipu3-cio2 1");
>>> + cio2_dm.add("ipu3-csi2 2");
>>> + cio2_dm.add("ipu3-cio2 2");
>>> + cio2_dm.add("ipu3-csi2 3");
>>> + cio2_dm.add("ipu3-cio2 3");
>>> +
>>> + cio2_ = enumerator->search(cio2_dm);
>>> + if (!cio2_)
>>> + return false;
>>> +
>>> + cio2_->acquire();
>>
>> Aha - never mind the above comments - We don't own the object to delete
>> it ... nor should it be deleted above :)
>>
>>> +
>>> + DeviceMatch imgu_dm("ipu3-imgu");
>>> + imgu_dm.add("ipu3-imgu 0");
>>> + imgu_dm.add("ipu3-imgu 0 input");
>>> + imgu_dm.add("ipu3-imgu 0 parameters");
>>> + imgu_dm.add("ipu3-imgu 0 output");
>>> + imgu_dm.add("ipu3-imgu 0 viewfinder");
>>> + imgu_dm.add("ipu3-imgu 0 3a stat");
>>> + imgu_dm.add("ipu3-imgu 1");
>>> + imgu_dm.add("ipu3-imgu 1 input");
>>> + imgu_dm.add("ipu3-imgu 1 parameters");
>>> + imgu_dm.add("ipu3-imgu 1 output");
>>> + imgu_dm.add("ipu3-imgu 1 viewfinder");
>>> + imgu_dm.add("ipu3-imgu 1 3a stat");
>>> +
>>> + imgu_ = enumerator->search(imgu_dm);
>>> + if (!imgu_) {
>>> + cio2_->release();
>>> + return false;
>>> + }
>>> +
>>> + imgu_->acquire();
>>> +
>>> + /*
>>> + * Disable all links that are enabled by default on CIO2, as camera
>>> + * creation enables all valid links it finds.
>>> + *
>>> + * Close the CIO2 media device after, as links are enabled and should
>>> + * not need to be changed after.
>>> + */
>>> + if (cio2_->open()) {
>>> + cio2_->release();
>>> + imgu_->release();
>>> + return false;
>>> + }
>>> + cio2_->disableLinks();
>>> +
>>> + numCameras_ = registerCameras();
>>> + LOG(Debug) << "\"Intel IPU3\" pipeline handler initialized with "
>>> + << numCameras_ << " cameras registered";
>>> +
>>> + cio2_->close();
>>> +
>>> + return true;
>>> +}
>>> +
>>> +/*
>>> + * Cameras are created associating an image sensor (represented by a
>>> + * media entity with function MEDIA_ENT_F_CAM_SENSOR) to one of the four
>>> + * CIO2 CSI-2 receivers.
>>> + *
>>> + * Cameras are here created and stored in the member field 'cameras_' map,
>>> + * indexed by the id of the CSI-2 receiver they are connected, to maintain
>>> + * an ordering that does not depend on the device enumeration order.
>>> + *
>>> + * The function returns the number of cameras found in the system.
>>> + */
>>> +unsigned int PipelineHandlerIPU3::registerCameras()
>>> +{
>>> + const std::vector<MediaEntity *> entities = cio2_->entities();
>>> + unsigned int numCameras = 0;
>>> + struct {
>>> + unsigned int id;
>>> + MediaEntity *csi2;
>>> + } csi2Receivers[] = {
>>> + { 0, cio2_->getEntityByName("ipu3-csi2 0") },
>>> + { 1, cio2_->getEntityByName("ipu3-csi2 1") },
>>> + { 2, cio2_->getEntityByName("ipu3-csi2 2") },
>>> + { 3, cio2_->getEntityByName("ipu3-csi2 3") },
>>> + };
>>> +
>>> + /*
>>> + * For each CSI-2 receiver on the IPU3, create a Camera if an
>>> + * image sensor is connected to it.
>>> + */
>>> + for (auto csi2Receiver : csi2Receivers) {
>>> + MediaEntity *csi2 = csi2Receiver.csi2;
>>> + unsigned int id = csi2Receiver.id;
>>> +
>>> + /* IPU3 CSI-2 receivers have a single sink pad. */
>>> + std::vector<MediaPad *> pads = csi2->pads();
>>> + MediaPad *sink;
>>> + for (MediaPad *pad : pads) {
>>> + if (!(pad->flags() & MEDIA_PAD_FL_SINK))
>>> + continue;
>>> +
>>> + sink = pad;
>>> + break;
>>> + }
>>> +
>>> + /* Verify that the receiver is connected to a sensor. */
>>> + std::vector<MediaLink *> links = sink->links();
>>> + if (links.empty())
>>> + continue;
>>> +
>>> + /*
>>> + * FIXME
>>> + * This supports creating a single camera per CSI-2 receiver.
>>> + */
>>> + for (MediaLink *link : links) {
>>> + MediaEntity *sensor = link->source()->entity();
>>> + if (!sensor)
>>> + continue;
>>> +
>>> + if (sensor->function() != MEDIA_ENT_F_CAM_SENSOR)
>>> + continue;
>>> +
>>> + /* Enable the link between sensor and CSI-2 receiver. */
>>> + if (link->setEnabled(true))
>>> + continue;
>>> +
>>> + /* Create the camera using the sensor name. */
>>> + std::size_t pos = sensor->name().find(" ");
>>> + std::string cameraName = sensor->name().substr(0, pos);
>>
>> I presume we'll have to make sure that this string is unique in the
>> system somewhere / somehow - but for now this is a good way to map the
>> cameras.
>>
>
> I cut off the bus address, if we keep that the name would be unique
> indeed. Otherwise yes, two sensors with the same name would give two
> cameras with the same name -> very bad. This needs to be changed, but
> instead of the bus address I would use the CSI-2 receiver index to
> form a unique name...
I'm not too worried here at the moment.
We can add in assertions to make sure we don't add a camera name which
already exists, and we can work out a unique naming later because it's
not going to be an ABI string. It's just a representation.
Might be worth adding a "Make sure camera names must be unique" task to
our task list though.
>
>>> +
>>> + cameras_[id] = new Camera(cameraName);
>>> +
>>> + LOG(Debug) << "Registered Camera[" << numCameras
>>> + << "] \"" << cameraName << "\""
>>> + << " connected to CSI-2 receiver " << id;
>>> +
>>> + numCameras++;
>>> + break;
>>> + }
>>> + }
>>> +
>>> + return numCameras;
>>> +}
>>> +
>>> +REGISTER_PIPELINE_HANDLER(PipelineHandlerIPU3);
>>> +
>>> +} /* namespace libcamera */
>>> diff --git a/src/libcamera/pipeline/ipu3/meson.build b/src/libcamera/pipeline/ipu3/meson.build
>>> new file mode 100644
>>> index 0000000..0ab766a
>>> --- /dev/null
>>> +++ b/src/libcamera/pipeline/ipu3/meson.build
>>> @@ -0,0 +1,3 @@
>>> +libcamera_sources += files([
>>> + 'ipu3.cpp',
>>> +])
>>> diff --git a/src/libcamera/pipeline/meson.build b/src/libcamera/pipeline/meson.build
>>> index 615ecd2..811c075 100644
>>> --- a/src/libcamera/pipeline/meson.build
>>> +++ b/src/libcamera/pipeline/meson.build
>>> @@ -1,3 +1,5 @@
>>> libcamera_sources += files([
>>> 'vimc.cpp',
>>> ])
>>> +
>>> +subdir('ipu3')
>>>
>>
>> --
>> Regards
>> --
>> Kieran
--
Regards
--
Kieran
More information about the libcamera-devel
mailing list