mirror of
https://github.com/meshtastic/firmware.git
synced 2026-08-01 19:08:59 -04:00
0eaad08735
* Fix T1000-E QMA6100P I2C probing * Refactored to make generic * address copilot comments * fix(i2c): address copilot comments on QMA6100P scanning and hardware init - ScanI2CTwoWire: gate bounded QMA6100P probing to addresses 0x12/0x13 only (Copilot comment: avoid blocking other I2C device detection on nRF52) Falls back to normal Wire probing for all other addresses, allowing BMM150 and other devices at overlapping addresses to be properly detected. - Nrf52Twim: suppress cppcheck redundantAssignment on ENABLE register write (intentional disable→configure→enable pattern for hardware safety; explicit disable ensures known state before configuration even if already disabled) Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com> * Nrf52Twim: gate on HAS_QMA6100P; suppress cppcheck redundantAssignment The TWIM helper is only used by the QMA6100P probing path (T1000-E, the sole board defining HAS_QMA6100P). It was gated only on ARCH_NRF52, so it compiled as dead code on every other nRF52 board and was analyzed by the rak4631 cppcheck job -- which failed on a redundantAssignment false positive. Gate the header and source on HAS_QMA6100P so the file is only built/analyzed where it is actually used. Also keep an inline suppression on the ENABLE re-assignment (a required volatile disable->reconfigure->enable sequence) for the case where the T1000-E build is run through cppcheck locally. Fixes the rak4631 cppcheck CI failure for PR #10713. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Ben Meadors <benmmeadors@gmail.com> Co-authored-by: nomdetom <nomdetom@protonmail.com> Co-authored-by: Tom <116762865+NomDeTom@users.noreply.github.com> Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>