Morning All,
I have a strange (and intermittent) issue with using the debug interface (SWD) on the MIMXRT1020-RTK.
It doesn't matter whether I use the on-board OpenSDA or an external J-Link. I am using the latest version of IAR Embedded Workbench.
The issue occurs when the line
CLOCK_SetMux(kCLOCK_PeriphMux, 0x0);
is executed in clock_config.c (which is NXP reference code - I haven't changed it).
After that line, the debugger can no longer communicate with the core, and the board needs to be power cycled.
Has anybody else seen the same effect?
If I comment out the line, everything seems to work fine (but I suspect there is some clock configuration difference that I haven't traced yet).
Screenshots of the offending line, and the IAR Embedded Workbench error attached.
It doesn't always occur, it seems to occur regularly and repeatedly, then suddenly is OK for days at a time...
Thanks,
Chris.
Hi Chris Cowdery
I'm not able to reproduce your issue, which demo are you using? also, did you have the default jumpers settings in the Evaluation board of the RT1020 (RT1020-EVK) ?
Regards
Jorge Alcala
Hi Jorge,
I started with 'freertos-generic.zip'. Although my project is quite modified from there (I have added a wireless stack).
I have altered the jumpers as I am powering the board via J2 (so J1 is connecting pins 1-2). I have also removed the SDRAM and Codec as I have soldered on an SPI wireless device instead. I have changed the jumpers to allow an external J-Link debugger to be used.
However I have seen this issue on a non-modified board straight from NXP.
Because it is random, I do not expect anyone to be able to reproduce it. But people may 'discover' they have the same issue.
Chris.
Hi Chris
Because I'm not able to reproduce the issue and that you see this issue in an Evaluation board, I believe the problem is with the software that you are using, it could be that you are running out of ram memory, or there is an error with the FreeRtos project that you created.
Regards
Jorge Alcala
Hi Jorge,
I will have to investigate further when it next occurs.
Hopefully in the meantime somebody else will see the same issue.
Regards,
Chris.