Registers GPIO3 and GPIO4 for secondary I2C bus

[Bajotumn]

Resolves #8417

Given pins 41 and 42 are used for GPS on this version it would seem using those pins, though drop in compatible with the V3 board, would render the GNSS socket unusable unless one used the inconvenient OLED pins on the forward edge of the board.

I will test this on a Heltec v3, RAK 4631, and a T-Beam as I have those, but a regression seems quite unlikely given this is a simple variant.h update.

🤝 Attestations

• I have tested that my proposed changes behave as described.
• I have tested that my proposed changes do not cause any obvious regressions on the following devices:
  • Heltec (Lora32) V3
  • LilyGo T-Deck
  • LilyGo T-Beam
  • RAK WisBlock 4631
  • Seeed Studio T-1000E tracker card
  • Other (please specify below)

[CLAassistant]

All committers have signed the CLA.

[fifieldt]

This probably needs WIRE_INTERFACES_COUNT to be set to 2

[Bajotumn]

This probably needs WIRE_INTERFACES_COUNT to be set to 2

It looks like this is implicitly set in configuration.h when I2C_SDA1 is defined.

firmware/src/configuration.h

Lines 303 to 311 in 1ac2382

	#ifndef WIRE_INTERFACES_COUNT
	// Officially an NRF52 macro
	// Repurposed cross-platform to identify devices using Wire1
	#if defined(I2C_SDA1) || defined(PIN_WIRE1_SDA)
	#define WIRE_INTERFACES_COUNT 2
	#elif HAS_WIRE
	#define WIRE_INTERFACES_COUNT 1
	#endif
	#endif

Happy to make this change if we think it should be explicitly defined though.

[Bajotumn]

And here's the testing setup

[b8b8]

Nice, thanks for adding this

[HarukiToreda]

I would suggest reviewing this one again, Heltec recently introduced this Touch expansion carrier and it is meant to fully integrate to the existing assigned pins. So it's best to first see what the i2c pin is used for it to avoid conflicts in plans already in the works

https://heltec.org/project/wifi-lora-32-v4-expansion-housing/

[b8b8]

they appear to be setup for sensor use and pass through to the back as expected (but reversed????)

[Bajotumn]

they appear to be setup for sensor use and pass through to the back as expected (but reversed????)

Very interesting. Good catch on the pass through to the external sockets.

I have one of these coming in and can verify once it arrives if SDA and SCL need swapping. It does look like the official Heltec code still has them defined as I do here. Time will tell.

[b8b8]

they appear to be setup for sensor use and pass through to the back as expected (but reversed????)

Very interesting. Good catch on the pass through to the external sockets.

I have one of these coming in and can verify once it arrives if SDA and SCL need swapping. It does look like the official Heltec code still has them defined as I do here. Time will tell.

Could just be a documentation error and nothing needs to change then. I dont think there is any sensor actually hooked up to the add-on board this would affect.

[HarukiToreda]

yeah probably doesn't need to change, but in the word of caution, it's good to make sure with the actual carrier, harder to change it once people have already soldered stuff to it and suddenly it switched from an update..

[Bajotumn]

yeah probably doesn't need to change, but in the word of caution, it's good to make sure with the actual carrier, harder to change it once people have already soldered stuff to it and suddenly it switched from an update..

So far, hopefully that's just me! 😀