Re: Pi Pico and Zephyr RTOS

On 12/01/2025 18:45, Brian Gregory wrote:

On 25/12/2024 10:02, mm0fmf wrote:

On 25/12/2024 00:30, wmartin wrote:

On 12/24/24 12:47, mm0fmf wrote:

Has anyone used the Zephyr RTOS on a Pi Pico or any other smallish system?
>
I'm writing something relatively simple for a Pico in C using the SDK. It doesn't need an RTOS, the old classic cyclic executive paradigm will work fine.
At one time I'd have used an 8bit CPU but why faff about when you can get something like a Pico for hardly any cost even if 2x Cortex M0+ seems overkill.

Well, you can have two threads, kinda, with the Pico without an RTOS...just split your work between the two cores. I just did a simple project using one core to do a com link to a pc for command/response, and the other core to do a real-time hardware controller chore. Made it painless...

>
That was my plan. The project has to do simple things like scan a 4x4 keyboard, light assorted LEDs and send strings over the UART at 38k4 to control something. I'd already decided that I would run all the UART send/receive on one core and run the control on the other core. That way the control core remains non-blocking and the UART core can block if needed.

 Doesn't the PICO SDK include some FreeRTOS examples?

It does. If You cant live without a pre-emptive multtaskking solution

 Do you really need Zephr RTOS rather than FreeRTOS?
 Note though, that I'm not sure if the examples have the ability to use both cores yet, I remember that initially they just allowed multi-threading using a single core on the PICO.
 If it was me I think I might just write my own RTOS that worked the same way I was originally taught to do multi-threading (using Dykstra's semaphores and not much else) rather than having to learn and understand how somebody else thought it should be done.
 

+1 for that.
--
I would rather have questions that cannot be answered...
...than to have answers that cannot be questioned
Richard Feynman