[libcamera-devel,12/13] gstreamer: Split completed request processing to a separate function

16358 diff mbox series

Simplify the task run function futher by moving the processing of
completed requests to a separate function. No functional change
intended, only increased readability.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
 src/gstreamer/gstlibcamerasrc.cpp | 125 +++++++++++++++++-------------
 1 file changed, 73 insertions(+), 52 deletions(-)

Hi Laurent,

Le vendredi 24 juin 2022 à 02:22 +0300, Laurent Pinchart a écrit :
> Simplify the task run function futher by moving the processing of
> completed requests to a separate function. No functional change
> intended, only increased readability.
> 
> Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> ---
>  src/gstreamer/gstlibcamerasrc.cpp | 125 +++++++++++++++++-------------
>  1 file changed, 73 insertions(+), 52 deletions(-)
> 
> diff --git a/src/gstreamer/gstlibcamerasrc.cpp b/src/gstreamer/gstlibcamerasrc.cpp
> index fb39d6093a3f..3feb87254916 100644
> --- a/src/gstreamer/gstlibcamerasrc.cpp
> +++ b/src/gstreamer/gstlibcamerasrc.cpp
> @@ -125,6 +125,7 @@ struct GstLibcameraSrcState {
>  
>  	int queueRequest();
>  	void requestCompleted(Request *request);
> +	int processRequest();
>  };
>  
>  struct _GstLibcameraSrc {
> @@ -241,6 +242,63 @@ GstLibcameraSrcState::requestCompleted(Request *request)
>  	gst_task_resume(src_->task);
>  }
>  
> +int GstLibcameraSrcState::processRequest()

Please comment that this must be called with steam lock held (which _run() holds
automatically).

> +{
> +	std::unique_ptr<RequestWrap> wrap;
> +
> +	{
> +		MutexLocker locker(lock_);
> +
> +		if (!completedRequests_.empty()) {
> +			wrap = std::move(completedRequests_.front());
> +			completedRequests_.pop();
> +		}
> +	}
> +
> +	if (!wrap)
> +		return -ENODATA;
> +
> +	GstFlowReturn ret = GST_FLOW_OK;
> +	gst_flow_combiner_reset(src_->flow_combiner);
> +
> +	for (GstPad *srcpad : srcpads_) {
> +		Stream *stream = gst_libcamera_pad_get_stream(srcpad);
> +		GstBuffer *buffer = wrap->detachBuffer(stream);
> +
> +		FrameBuffer *fb = gst_libcamera_buffer_get_frame_buffer(buffer);
> +
> +		if (wrap->pts_) {
> +			GST_BUFFER_PTS(buffer) = wrap->pts_;
> +			gst_libcamera_pad_set_latency(srcpad, wrap->latency_);
> +		} else {
> +			GST_BUFFER_PTS(buffer) = 0;
> +		}
> +
> +		GST_BUFFER_OFFSET(buffer) = fb->metadata().sequence;
> +		GST_BUFFER_OFFSET_END(buffer) = fb->metadata().sequence;
> +
> +		ret = gst_pad_push(srcpad, buffer);
> +		ret = gst_flow_combiner_update_pad_flow(src_->flow_combiner,
> +							srcpad, ret);
> +	}
> +
> +	if (ret != GST_FLOW_OK) {
> +		if (ret == GST_FLOW_EOS) {
> +			g_autoptr(GstEvent) eos = gst_event_new_eos();
> +			guint32 seqnum = gst_util_seqnum_next();
> +			gst_event_set_seqnum(eos, seqnum);
> +			for (GstPad *srcpad : srcpads_)
> +				gst_pad_push_event(srcpad, gst_event_ref(eos));
> +		} else if (ret != GST_FLOW_FLUSHING) {
> +			GST_ELEMENT_FLOW_ERROR(src_, ret);
> +		}
> +
> +		return -EPIPE;
> +	}
> +
> +	return 0;
> +}
> +
>  static bool
>  gst_libcamera_src_open(GstLibcameraSrc *self)
>  {
> @@ -308,8 +366,13 @@ gst_libcamera_src_task_run(gpointer user_data)
>  	GstLibcameraSrc *self = GST_LIBCAMERA_SRC(user_data);
>  	GstLibcameraSrcState *state = self->state;
>  
> -	int err = state->queueRequest();
> -	if (err == -ENOMEM) {
> +	/*
> +	 * Create and queue one request. If no buffers are available the
> +	 * function returns -ENOBUFS, which we ignore here as that's not a
> +	 * fatal error.
> +	 */
> +	int ret = state->queueRequest();
> +	if (ret == -ENOMEM) {
>  		GST_ELEMENT_ERROR(self, RESOURCE, NO_SPACE_LEFT,
>  				  ("Failed to allocate request for camera '%s'.",
>  				   state->cam_->id().c_str()),
> @@ -318,58 +381,16 @@ gst_libcamera_src_task_run(gpointer user_data)
>  		return;
>  	}
>  
> -	std::unique_ptr<RequestWrap> wrap;
> -
> -	{
> -		MutexLocker locker(state->lock_);
> -
> -		if (!state->completedRequests_.empty()) {
> -			wrap = std::move(state->completedRequests_.front());
> -			state->completedRequests_.pop();
> -		}
> -	}
> -
> -	if (!wrap) {
> -		gst_task_pause(self->task);
> -		return;
> -	}
> -
> -	GstFlowReturn ret = GST_FLOW_OK;
> -	gst_flow_combiner_reset(self->flow_combiner);
> -
> -	for (GstPad *srcpad : state->srcpads_) {
> -		Stream *stream = gst_libcamera_pad_get_stream(srcpad);
> -		GstBuffer *buffer = wrap->detachBuffer(stream);
> -
> -		FrameBuffer *fb = gst_libcamera_buffer_get_frame_buffer(buffer);
> -
> -		if (wrap->pts_) {
> -			GST_BUFFER_PTS(buffer) = wrap->pts_;
> -			gst_libcamera_pad_set_latency(srcpad, wrap->latency_);
> -		} else {
> -			GST_BUFFER_PTS(buffer) = 0;
> -		}
> -
> -		GST_BUFFER_OFFSET(buffer) = fb->metadata().sequence;
> -		GST_BUFFER_OFFSET_END(buffer) = fb->metadata().sequence;
> -
> -		ret = gst_pad_push(srcpad, buffer);
> -		ret = gst_flow_combiner_update_pad_flow(self->flow_combiner,
> -							srcpad, ret);
> -	}
> -
> -	if (ret != GST_FLOW_OK) {
> -		if (ret == GST_FLOW_EOS) {
> -			g_autoptr(GstEvent) eos = gst_event_new_eos();
> -			guint32 seqnum = gst_util_seqnum_next();
> -			gst_event_set_seqnum(eos, seqnum);
> -			for (GstPad *srcpad : state->srcpads_)
> -				gst_pad_push_event(srcpad, gst_event_ref(eos));
> -		} else if (ret != GST_FLOW_FLUSHING) {
> -			GST_ELEMENT_FLOW_ERROR(self, ret);
> -		}
> +	/* Process one completed request, if available. */
> +	ret = state->processRequest();
> +	switch (ret) {
> +	case -EPIPE:
>  		gst_task_stop(self->task);
>  		return;
> +
> +	case -ENODATA:
> +		gst_task_pause(self->task);
> +		return;
>  	}

Looks good to me.

Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>

>  
>  	/*