STM32L4 sometimes hangs after NVIC_SystemReset()

[PGoud]

About once over 10, the MCU seems to hang after NVIC_SystemReset() call. Notice that:

• The MCU restarts correctly after poweron reset,
• NRESET set to 0 then 1 does NOT restart the MCU
• When MCU is hung, the NRESET Level is correctly UP (thanks to internal pull-up)

[Tesla DeLorean]

Is this a custom board? Is anything else driving NRESET?​

[PGoud]

Yes it is a custom board.

NRESET is only connected through a 1nF capacitance toward 0v.

But notice that when the MCU is hung, the level at NRESET is 3.2 V

[PGoud]

New information.

I can systematically set my MCU in this "hung" mode after :

• Init of my board (HAL, Clock etc ...) and GPIOs THEN NVIC_SystemReset()

BUT this is systematically "hanging" ONLY if

• my software has just been reprogrammed and automatically restarted
• my board is attach to the programming STLink (obviously it has just been reprogrammed ;O),

This with a production (not debug) software that set :

• JTMS/JTCK/JTDI/JTDO in MODE ANALOG / NO PULL
• JNRESET in MODE INPUT / PULL UP

[waclawek.jan]

STLink connected to NRST?

JW

[PGoud]

Sometimes.

The problem can occur connected or not connected.

But for a while I thougt it was easyer to produce when connected and software has just been programmed.