RPMsg in thingy53 with Segger J-Link mini Edu Not showing the messages in the console

[neuberfran]

Hi,

I stay trying this project github.com/.../rpmsg_playground
based on this video: www.youtube.com/watch
(openAMP/RPMsg in nordic thingy:53(nrf5340 dk)

Everything compiles and flashes OK, but the RPMsg messages from the Cores don't come out in the Console. I've used JLinkExe, JLinkExe -if SWD
(Both with h and g option combined after connect and speed) and also combined with JLinkRTTClient opened in another terminal

I stay using JLINK MINI EDU By SEGGER

Notes:

1) i stay using zephyr 3.3 and NCS

2) In github link I have several photos (like the one below) that demonstrate my issue:

https://github.com/neuberfran/rpmsg_playground/blob/main/photo04.png

3) I already did a blink led on this board now (using Zephyr 3.3) and it worked. Then I used the command: nrfjprog --eraseall to clean up.

How to solve?

?

[Sigurd Hellesvik]

Hi,

Which version of the nRF Connect SDK do you use?

I already did a blink led on this board now

Before running this sample not in our nRF Connect SDK, can you log from the Thingy53 using other means (using nRF Connect SDK samples)?
For example over USB or over RTT

Regards,
Sigurd Hellesvik

[neuberfran]

You are sure. I have now checked the links and the answer is No. Apart from the blink led example, I didn't make any other examples and therefore, I didn't delve into the use of USB and RTT. In fact, in my RPMsg/openAMP application (reason for this post) I have this file below all commented out (this file was also excluded in the model video) because, if the file is not commented out, I get the following error:

https://drive.google.com/file/d/1pzxOuqfDOxh5ymEAMJRu1dsQPctthzLy/view?usp=sharing

gist.github.com/.../914d74c63953be16a049e7bdb3083715

[neuberfran]

"My understanding is that you've had issues getting the log from the netcore of a 53?"

Thanks for taking this case. In reality, it is now more in-depth learning of traffic lights. Really, I haven't had good logs on this project. I don't know if due to misuse (or use in this case) of USB_CDC. I'm using RTT just like the original video. But (logs) shouldn't be a problem,
because I invested in a SEGGER Edu Mini.

"Did you copy the project made in the video, or did you just try to follow along on some steps mentioned in it? If so, could it be that you've skipped a step?"

1) Yes
2) More or less. The file below is the only one that raised a doubt

https://github.com/neuberfran/rpmsg_playground/blob/main/aci/CMakeLists.txt

https://postimg.cc/Xr0ttqjZ

As you can see in the image above, if I change line 4 (for example) to: SOURCE_DIR ${CMAKE_CURRENT_LIST_DIR}/..../cpunet I don't get any errors. this is intriguing

Finally, I would like you to prove to me that I only need to comment this line now, because my project is newer than the one in the video and in the video the author used an older version of ncs and zephyr
If that's the case, of course, there's no problem and there's nothing you can do about it.

The project is on my github if you stay trying do in your 53

Tks

[Elfving]

neuberfran said:

As you can see in the image above, if I change line 4 (for example) to: SOURCE_DIR ${CMAKE_CURRENT_LIST_DIR}/..../cpunet I don't get any errors. this is intriguing

Maybe I am overlooking something obvious, but I am not really seeing the difference between the image and the github link.

Are you saying that when you added a few more dots it worked? Or that when you've commented it out it works?

neuberfran said:

because my project is newer than the one in the video and in the video the author used an older version of ncs and zephyr
If that's the case, of course, there's no problem and there's nothing you can do about it.

I guess you could downgrade to the NCS & Zephyr version that they used. Do you know which one that could be?

Regards,

Elfving

[neuberfran]

https://postimg.cc/9DB0cDKR

Yes. I make changes to this line or another in this file and the compilation does not show any error(s)

I guess you could downgrade to the NCS & Zephyr version that they used. Do you know which one that could be?

I don't know which version was used in the video

[Elfving]

I see.

neuberfran said:

Yes. I make changes to this line or another in this file and the compilation does not show any error(s)

What sort of changes? When you comment it out or when you add some dots to the path cmakelists was referring to?

Regards,

Elfving

[neuberfran]

Ok bro.We can forget about the question about the cmakelists file and its changes having no effect.We can focus on the main issue from now on: out the line k_sem_take(&bound_sem, K_FOREVER); (in two places) in cpuappNet's main.c code: 

I honestly don't know the versions of zephyr and ncs used in the video and I don't know if the changes in these versions are what cause this change of having to comment lines.

[Sigurd Hellesvik]

Hi,

I am back

neuberfran said:

github.com/.../quote]

I cloned yoir project and am testing it on a nRF5340DK.
Here, I get the following log:

Is this expected?

I struggle to understand the conversation below.
Can you try to shortly sum up what the error is and what you expect to see when it works?

Regards,
Sigurd Hellesvik