Add receive-only same-image display, SPI, modem/IQ-cache and scoped-priority experiments with checkpointed collectors. Keep rejected clock-phase correction and elevated-priority experiments out of production.
Publish sanitized timing data, Full Companion/Bluetooth/USB checks and redraw measurements with their limits. Wire the new behavioral checks into CI.
Reuse the confirmed LoRa modem only inside acknowledged warm continuous-RX hops and revoke that capability before modem initialization or after failures.
Opt the tested T-Deck into IQ workaround reuse and an exact 8,888,888 Hz SPI setting. Keep the existing BUSY, packet-ownership and recovery checks; other boards retain their modem-cache default.
Render into a PSRAM canvas and release the shared SPI bus between bounded TFT strips. Service only the radio driver at cooperative checkpoints and use the buffered radio HAL, preserving packet and transmit guards.
Cover pixel parity, transaction boundaries, allocation failure, partial updates and sleep/wake behavior with native tests.
Use the committed node credential for setup WiFi, WebConfig and browser OTA. Add persistent ESP32 WiFi Companion management credentials without changing preferences layouts, revoke stale browser sessions on rotation, retain scoped save receipts, and coordinate credential changes with flash uploads. Keep external-service credentials and trusted TCP protocol compatibility independent. Update live join QR codes, failure handling and documentation.
Validated 51 native cases, 158 related host and browser methods, and serial Full V4 Companion, Station G3 Repeater, ESP32-C6 Room Server and V4 Sensor firmware builds.
Deduplicate validation and preserve the published preference bytes and transactional saves. Correct Ikoka Handheld Companion flash bounds and document QSPI Room migration. Qualify STM32 and RAK4631 at the full 1.18.0.1 release version length.
Integrate fractional Companion timezones and conditional Bluetooth pages while preserving the existing preference byte layout, transactional saves, and Full transport selector. Add migration, rollback, invalid-value, and UI profile regressions.
/api/login locked every client for 30 s after five failures from anyone.
Anyone who could reach the login page could keep the owner locked out by
failing on purpose. If no login succeeded after a lockout, the lock also
returned from 2^31 to 2^32 ms (about 24.9 to 49.7 days) later, because the
deadline was never cleared and was compared as a signed difference.
Count failures per remote address instead. Five failures lock that address
for 30 s, as before. A shared budget caps failures from all addresses
together, so changing address does not add guesses: it holds six failures
and refills one every 6 s, the old global rate, so one failing address can
never empty it on its own. While it is empty every client waits, as before.
Lock times are measured as elapsed time.
The policy is a pure header, WebConfigLoginThrottle.h.
test_webconfig_login_throttle.py runs it directly (isolation, boundaries,
one-hour rate bounds, millis() half-wrap, saturation, eviction) and runs the
real handleLogin with ArduinoJson under ASan/UBSan. Putting every client on
one key, as the old code did, fails the handler test.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Restart the retry interval without treating queue cleanup as completed USB transmission. Cover cancellation journals, fresh recovery deadlines, real progress and rollover, with historical policy and runtime negative controls.
Keep rejected power-saving requests from changing radio2's saved RX policy or timings. Exercise actual CLI, scan restoration and calibration sleep guards with semantic negative controls.
mikecarper
nullrouten0andClaude Opus 5.5