But as per the other forum solutions, to extract embedded data we need to set embedded_metadata_height = “1”. But while applying this, I’m facing some stuck in streaming.
Please check the output while I had set the height as 1081 and embedded data as 0. How to extract the embedded data from the above 1920*1081 buffer or Is there any files which I can take as reference?. Also, Is there any issues in the ISP binary which I’m using?
The embedded data line shouldn’t include to the height. Just add correct line for the embedded_metadata_height and keep report the output exclude the embedded data line.
/*
* WARNING: frmfmt ordering need to match mode definition in
* device tree!
*/
static const struct camera_common_frmfmt ox03a10_frmfmt[] = {
{{1920, 1080}, ox03a10_20fps, 1, 0, OX03A10_MODE_1920X1080_20FPS},
/* Add modes with no device tree support after below */
};
While executing the below pipeline in order to get streaming, I’m getting a stuck in the streaming.
Can you suggest me what might be the reason for the wrong embedded data type?. We’re getting the stream from an external ISP. Should we need to alter the ISP Image in order to resolve this issue?
Can you help me to figure out what’s the current value of embedded data type from the trace what I shared.
Also, to make it clear the sensor vendor asked about what does this Embedded Data Type=0x12 means?. Is that the Generic 8 bit long packet data types mentioned in the below document?(page 73)
static const struct camera_common_frmfmt ox03a10_frmfmt[] = {
{{1920, 1080}, ox03a10_20fps, 1, 0, OX03A10_MODE_1920X1080_20FPS},
/* Add modes with no device tree support after below */
};
I also wanted to add one more observation. When I tried to set embedded_data_height as 0, I’m getting streaming without any stuck. But still some irregularities in the streaming.
static const struct camera_common_frmfmt ox03a10_frmfmt[] = {
{{1920, 1080}, ox03a10_20fps, 1, 0, OX03A10_MODE_1920X1080_20FPS},
/* Add modes with no device tree support after below */
};
The datatype expected for Embedded Data is 0x12 and for Image data it needs to be 0x1E If I’m planning to collect Embedded data and Image Data(YUV422 8-bit) simultaneously, right?