We're using a PICO-IMX7 SOM (based on an iMX7D) and have discovered a problem. At 500ms after power-on, we experience a brief (around 2ms) spike on one of our outbound GPIO lines (see attached image). We assumed that the problem would be some rogue initialization code in the SPL but it actually seems to happen before any SPL code runs. We've tested this by using a dummy SPL that does nothing but loop forever (annotated hexdump attached) and the spike still occurs.
As it turns out, this spike can sometimes lead to a large, potentially damaging, current spike which means we have to get to the bottom of this.
The curious thing is that it doesn't happen until 500ms - surely by then the SPL should already be running? The pad that we see the spike on is SAI1_TXFS.
Does anyone have any idea about what could be causing this?
Hi Martin
PICO-IMX7 SOM is third party board developed and supported directly by Technexion
https://www.technexion.com/products/system-on-modules/pico/pico-imx7/
May be recommended to apply to vendor support:
https://www.technexion.com/support/
>>The curious thing is that it doesn't happen until 500ms - surely by
>then the SPL should already be running?
one can try to check if it is caused by some custom code by setting BOOT_MODE to "Serial Downloader"
(so no customer code was running) described in sect.6.6.2.4 Serial Downloader i.MX 7Dual Applications Processor Reference Manual
Best regards
igor
Thank you @igorpadykov for the response. Trying it in serial download mode is a great idea.
However, I'd be fascinated to know if there's anything in the IPL that could cause this. Is the source available? If not, is it feasible that it can take 500ms before the SPL code is executed?
Thank you!
> Is the source available?
I believe you should ask them from vendor - Technexion. This board is not supported by NXP
and for sources one can look at Technexion resources:
https://github.com/TechNexion-customization
Best regards
igor