camera2_frame_queue_dst_ops Struct Reference
camera2_frame_queue_dst_ops Struct Reference
#include <
camera2.h
>
Data Fields |
|
int(* | dequeue_frame )(const struct camera2_frame_queue_dst_ops *q, size_t entries, size_t data_bytes, camera_metadata_t **buffer) |
int(* | cancel_frame )(const struct camera2_frame_queue_dst_ops *q, camera_metadata_t *buffer) |
int(* | enqueue_frame )(const struct camera2_frame_queue_dst_ops *q, camera_metadata_t *buffer) |
Detailed Description
Frame output queue protocol:
The framework holds the queue and its contents. At start, the queue is empty.
- When the device is ready to fill an output metadata frame, it must dequeue a metadata buffer of the required size.
- It should then fill the metadata buffer, and place it on the frame queue using enqueue_frame. The framework takes ownership of the frame.
- In case of an error, a request to flush the pipeline, or shutdown, the device must return any affected dequeued frames to the framework by calling cancel_frame.
Field Documentation
int(* cancel_frame)(const struct camera2_frame_queue_dst_ops *q, camera_metadata_t *buffer) |
int(* dequeue_frame)(const struct camera2_frame_queue_dst_ops *q, size_t entries, size_t data_bytes, camera_metadata_t **buffer) |
Get an empty metadata buffer to fill from the framework. The new metadata buffer will have room for entries number of metadata entries, plus data_bytes worth of extra storage. Frames dequeued here must be returned to the framework with either cancel_frame or enqueue_frame.
int(* enqueue_frame)(const struct camera2_frame_queue_dst_ops *q, camera_metadata_t *buffer) |
The documentation for this struct was generated from the following file:
- hardware/libhardware/include/hardware/ camera2.h