Good day Steve,
I apologize. Due to the incestive reasons, we were quarantined. Alas,
all work is postponed for at least a week.
[...]
- entity 47: ipu1_csi1 (3 pads, 4 links)
type V4L2 subdev subtype Unknown flags 0
device node name /dev/v4l-subdev5
pad0: Sink
[fmt:SRGGB10_1X10/1920x1080@1/30 field:none
colorspace:srgb xfer:srgb ycbcr:601 quantization:full-range
crop.bounds:(0,0)/1920x1080
crop:(0,0)/640x480
compose.bounds:(0,0)/640x480
compose:(0,0)/640x480]
<- "imx6-mipi-csi2":2 [ENABLED]
This is virtual channel 1 output from the imx6-mipi-csi2 receiver. Make
sure the imx219 is indeed transmitting on virtual channel 1.
Khm... I belive my schematechnics. Electrical connection present.
Osciloscope show clock and data if I write stream enable command via I2C
to IMX219. Does it make sense to do additional checks? What kind?
- entity 51: ipu1_csi1 capture (1 pad, 1 link)
type Node subtype V4L flags 0
device node name /dev/video3
pad0: Sink
<- "ipu1_csi1":2 [ENABLED]
What is the format on /dev/video3 ? Please provide output from 'v4l2-ctl
-d3 -V'
Alas, this requires physical access to the board.
I think fakesink must got Bayer format
(as specific /dev/null for v4l2).
Can you try simply:
localhost ~ # v4l2-ctl -d3 --stream-mmap
and let me know if you receive frames.
Thanks. So far, I have not had to deal with the Linux media subsystem. I
only understand the design and hierarchy of drivers. Obviously, it is
far from always successful.
