-
Notifications
You must be signed in to change notification settings - Fork 69
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Working on IMX214 on RockPro64- no stream #67
Comments
please report the logs from:
(Capture 100 frames from the first camera) |
The output is
and it comes to a halt (and the camera makes a strange click noise once)... what I could also see is this on the kernel log
|
Seems like a fault setting the controls on the subdevice gets things upset? I'm not sure why that causes buffers to stop queueing though. Is there a patch that shows any changes you've made to the imx214 ? |
I made some progress... the media-ctl/video-ctl stuff now works for capturing images with the original imx214 kernel module, but it of course fails with libcamera, as it is incompatible... let me paste the working DTS for MIPI Port 0 (the one closer to the network interface) first:
Will post the output from the patched imx214 asap... PS: so it seems my "fix" to the driver repair broke it 🤣 |
While I was working on the the IMX214 already last year with limited success, I picked it up again and I need some buddies to discuss the problem I'm facing.
As it seems it is a mixture of device tree problems and problems with the library itself (probably). There is no forum to collect all the findings I made so far, so this is an attempt to do so and make the IMX214 finally work with libcamera - and to provide a working solution that hopefully makes use of libcamera as well.
To start off, here's what I did:
I originally had some trouble setting up the clocks, but those seem to be fine thx to a pro from the Discord forum. According to the specs there is GPIO2_B4 and GPIO0_B4 for the PowerDown-Pins, depending on the MIPI port used, but I can't really make sense out of them.
Anyway, here's the issue: the libcamera works and finds the camera, but stream mapping is broken (it gets stuck). Similarly lc_compliance gets stuck already at the first test. There is also this media-ctl sequence, which should get some data out at some point:
and ultimately
fails with "broken pipe" or "invalid device" or something similar. So it seems that libcamera has the same problem with the raw stream capture by media-ctl/video-ctl..., and I can't get rid of the feeling that I'm missing something somewhere. There was a conversation on the DietPi forum about this, but no real working solution at the end either...
I have a similar setup for the OV13850 as well, for which I also already patched the devicetree and libcamera - that's a slightly different story, but the outcome is the same... will try to summarize this in an other submission...
The text was updated successfully, but these errors were encountered: