search

zephyrproject-rtos / zephyr

Primary Git Repository for the Zephyr Project. Zephyr is a new generation, scalable, optimized, secure RTOS for multiple hardware architectures.
https://docs.zephyrproject.org
Apache License 2.0
10.33k stars 6.33k forks source link

SPI on STM32U5 mcu get stuck after 50f64eaeba commit (#63173) #66486

Closed avisconti closed 8 months ago

avisconti commented 8 months ago

After #63173 was merged I started to see an issue on my sensortile_box_pro board, based on stm32u5 mcu. In particular, reverting the 50f64eaeba commit seems to fix the issue. Looking at that commit I found out that, since stm32u5 mcu is based on stm32h7 spi controller, it is impacted by this change and it gets stuck into spi_stm32_shift_fifo() routine.

erwango commented 8 months ago

Closing as this is the same issue as https://github.com/zephyrproject-rtos/zephyr/issues/66326. But don't hesitate to share in this issue if you have a simple way to reproduce the issue.

avisconti commented 8 months ago

Closing as this is the same issue as #66326. But don't hesitate to share in this issue if you have a simple way to reproduce the issue.

@erwango I actually saw issue #66326 but I'm not sure it is the same one (it might be). In my case it gets stuck in a different routine.

dgastonochoa commented 8 months ago

Closing as this is the same issue as #66326. But don't hesitate to share in this issue if you have a simple way to reproduce the issue.

@erwango I actually saw issue #66326 but I'm not sure it is the same one (it might be). In my case it gets stuck in a different routine.

I have opened a PR that (at least in my setup) fixes the problem: #66503.

I'm not sure if the issue is the exact same one as you mention; it would be great if you could test it on your side to be 100% sure.

avisconti commented 8 months ago

Closing as this is the same issue as #66326. But don't hesitate to share in this issue if you have a simple way to reproduce the issue.

@erwango I actually saw issue #66326 but I'm not sure it is the same one (it might be). In my case it gets stuck in a different routine.

I have opened a PR that (at least in my setup) fixes the problem: #66503.

I'm not sure if the issue is the exact same one as you mention; it would be great if you could test it on your side to be 100% sure.

Yes, it seems to fix this issue as well. Thanks very much