Skip to content

Add SuperMini nRF52840 (and nice!nano / nRFMicro) to test matrix #224

Description

@zackees

Background

FastLED PR FastLED/FastLED#2445 (closes FastLED/FastLED#2422) adds FastPin variant blocks for three community nRF52840 boards exposed by the pdcook/nRFMicro-Arduino-Core BSP:

  • TARGET_SUPERMINI_NRF52840SuperMini nRF52840
  • TARGET_NICE_NANO_V2nice!nano v2 (ProMicro footprint)
  • TARGET_NRFMICROnRFMicro

The PR also adds ci/boards.py entries (supermini_nrf52840, nice_nano_nrf52840, nrfmicro_nrf52840) and a per-board GitHub Actions workflow (build_nrf52840_supermini.yml) on the FastLED side.

Ask

To keep board coverage consistent between the FastLED repo and fbuild's own regression matrix, please:

  1. Add supermini_nrf52840 (and ideally nice_nano_nrf52840 + nrfmicro_nrf52840) to whatever board-set fbuild iterates over for its self-tests.
  2. Surface a build badge (or extend the existing nRF52 badge group) on the fbuild README so the new board's status is visible alongside the other nRF52 boards already covered (e.g. nrf52840_dk, xiaoblesense).

Notes

  • All three boards share the nordicnrf52 PlatformIO platform with framework-arduinoadafruitnrf52@^1.10601.0 and board_build.core=nRF5. The new boards.py entries reuse nrf52840_dk_adafruit as the underlying real board, with the variant block selected via -DTARGET_*.
  • Local testing currently fails inside the Adafruit BSP cores/nRF5/utility/debug.cpp (ARDUINO_BSP_VERSION "too many decimal points" error) on GCC 15.2.1 — this is a pre-existing BSP/toolchain mismatch independent of FastLED's variant code. If fbuild's CI uses a different toolchain version this may not reproduce; otherwise the fix likely belongs in the BSP package pin.

cc @zackees

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions