Hi everyone, sorry for necroposting, I haven't been able to look at this project for months on end, partly due to repeated failed attempts to amend this issue.
I eventually ended up uploading the project to GitLab: https://gitlab.com/lachender-mann/rp2350-868-lora-board, so if you want to take a look at it yourself, you're more than welcome to do so.
Anyhow, I've ended up cutting the traces that went from the RP MCU flash interface to the SX radio IC, and that initially worked for a while, meaning that the MCU would successfully boot and not immediately crash, so that I could at least attempt a debug session with it. I'm attaching the pictures of the cut traces below, for reference. You will notice that some of them have been cut twice, and the reason I've done that is because after the first cut I didn't want to risk leaving a long stub trace going out some of the pins.
In this arrangement, SD0 to SD3 and the CLK line for the flash are not tied to anything, and thus shouldn't cause issues.
Nevertheless, I'm still encountering issues in normal execution of the code, and cannot trace these crashes to anything that makes sense to me. With the minimal reproducible example I've pasted in the OP, now what happens in ozone is that code will successfully upload, but will eventually immediately crash, with the PC set to 0x2F8 (inside bootrom)What's more, I'm still not able to infer any additional information regarding the fault issues, see Ozone's crash report:I've tried the following workarounds so far:
- disabling interrupts as soon as I enter main(), and loop there
- do the same, but within the CRT routine
- step by step until main
But none have worked consistently so far, the first two won't avoid the issue and the third one will successfully enter the main() routine, but as soon as the execution is resumed, the same crash occurs.
Any suggestions on how to proceed? Thanks!
I eventually ended up uploading the project to GitLab: https://gitlab.com/lachender-mann/rp2350-868-lora-board, so if you want to take a look at it yourself, you're more than welcome to do so.
Anyhow, I've ended up cutting the traces that went from the RP MCU flash interface to the SX radio IC, and that initially worked for a while, meaning that the MCU would successfully boot and not immediately crash, so that I could at least attempt a debug session with it. I'm attaching the pictures of the cut traces below, for reference. You will notice that some of them have been cut twice, and the reason I've done that is because after the first cut I didn't want to risk leaving a long stub trace going out some of the pins.
Nevertheless, I'm still encountering issues in normal execution of the code, and cannot trace these crashes to anything that makes sense to me. With the minimal reproducible example I've pasted in the OP, now what happens in ozone is that code will successfully upload, but will eventually immediately crash, with the PC set to 0x2F8 (inside bootrom)
- disabling interrupts as soon as I enter main(), and loop there
- do the same, but within the CRT routine
- step by step until main
But none have worked consistently so far, the first two won't avoid the issue and the third one will successfully enter the main() routine, but as soon as the execution is resumed, the same crash occurs.
Any suggestions on how to proceed? Thanks!
Statistics: Posted by bru3s — Sat Nov 29, 2025 9:24 am