How to use target reset functionality?

[recursivenomad]

How to use target reset functionality?

[P33M]

Fixed in #123

[lurch]

I believe that the intention is that you'd wire PROBE_PIN_RESET up to the RUN pin on the target board that you're debugging? I.e. that's why it's a "hardware reset" rather than a "software reset".

(Ahh, although re-reading your original post, perhaps I've misunderstood what it is that you're asking?)

[recursivenomad]

Correct, I've been experimenting with wiring to both the RUN pin and the 3V3_EN pin. My trouble is understanding how to use this functionality once recompiling and uploading the modified picoprobe firmware - It doesn't appear to be hardware-resetting the target Pico (although I'll admit I haven't scoped it with a logic analyzer yet).

I'm curious if there's a specific command I can send through GDB/OpenOCD to manually activate the reset line functionality? Looking for documentation on how to use this feature after it's already been wired up.

[lurch]

This is just a guess, but I wonder if that might be part of the old "custom picoprobe interface", and not yet part of the (now-default) "standard cmsis-dap interface" ? 🤷

Ahh... looks like 0761424 is the commit that hooks it into the "standard cmsis-dap interface", so is probably a more-useful starting point than #10 ?

[P33M]

Can you try the options listed in the OpenOCD user guide? https://openocd.org/doc/html/Reset-Configuration.html

[recursivenomad]

Okay, finally found the time to test everything with a logic analyzer attached 🙂 Bottom line, none of the configuration options I tried from that OpenOCD page managed to produce a low signal on PROBE_PIN_RESET (I was monitoring the probe's pins 0 and 1 for good measure).

That said, I can confirm that sending monitor cmsis-dap info through GDB returns ... nRESET = 1, which is a promising start.

Adding "reset_config srst_only" to "openOCDLaunchCommands" in my lauch.json changes the output of monitor reset_config, but still no low signal on PROBE_PIN_RESET.

And same goes for manually sending monitor adapter assert srst via GDB. No signal.

Setting monitor debug 3 and viewing the OpenOCD output shows that OpenOCD believes SRST is asserted during reset; but again, no change in signal.

I looked into the target configuration file and did find this in OpenOCD's .../target/rp2040.cfg:

	# srst does not exist; use SYSRESETREQ to perform a soft reset
	$_TARGETNAME_0 cortex_m reset_config sysresetreq

However, I don't believe this is related as OpenOCD believes it is asserting and releasing SRST during reset when it is enabled.