Commit Graph
1228 Commits
Author SHA1 Message Date
tthandClaude Fable 5 34bb9a1aa0 fix(repeater): drobnosti z review antispamu (žádný high/med, jen LOW)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-09-06 00:07:52 +02:00
tthandClaude Fable 5 ce4b7bd343 ci: opravit nRF52 buildy na GH (lokální symlink leakl do trackovaného configu)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-09-06 00:07:38 +02:00
tthandClaude Fable 5 69e321221a feat(repeater): fairness grid viditelnost + runtime cap/refill tuning
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-09-05 23:39:04 +02:00
tthandClaude Fable 5 7e21b6d521 feat(repeater): antispam — FairnessLimiter (port #1502) + kanálový deny list
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-09-05 21:35:36 +02:00
tthandClaude Opus 5 82a74318c4 Merge remote-tracking branch 'origin/dev' into main
58 commitů upstreamu nad v1.17.1. Šest konfliktů, všechny ručně:

- `src/helpers/radiolib/RadioLibWrappers.cpp` — OBOJE. Upstream `aa229245`
  přidává `idle()` před `setOutputPower()` na LR2021 (tam jsou to standby-only
  příkazy a přijímač se sám nevrátí do Rx); naše část si dál drží návratový
  status, aby „rádio to odmítlo" a „rádio to změnilo" nevypadaly z konzole
  stejně. Nezávislé věci, obě zůstávají.
- `src/helpers/CommonCLI.cpp` — VERZE UPSTREAMU. Celý blok `radio`,
  `radio.rxgain` a `radio.fem.*` byl čistě upstreamový kód (diff proti merge
  base v té oblasti je prázdný) a upstream ho přestěhoval do nového
  `CommonRadioPrefs`; FEM příkazy zrušil ve prospěch per-variant klíčů
  `fem_rxgain`/`fem_txgain`.
- `examples/simple_repeater/MyMesh.h` — NAŠE + úklid. `static_assert` na
  64bajtový wire formát `RepeaterStats` zůstává, `MAX_CLIENTS` odsud mizí:
  `1a8d860d` ho centralizoval do `helpers/ClientACL.h`, který je tu už
  includovaný.
- `examples/simple_repeater/MyMesh.cpp` — NAŠE. Konflikt byl falešný (upstream
  ten řádek nemění), jde o naše dva countery `tx_start_fail`/`tx_timeout`.
- `examples/companion_radio/ui-new/UITask.cpp` — OBOJE. Upstream přidal větve
  pro `UI_SHOW_CLOCK`, my podmínku `_task->isBluetoothEnabled()`, aby displej
  nenabízel PIN k párování, které není jak navázat.
- `platformio.ini` — OBOJE. Vedle našeho symlinku na lokální nRF52 framework
  přibyl upstreamový `toolchain-gccarmnoneeabi@^1.140201.0`.

Navíc: `[env:native_asan]` musel dostat `DynamicConfigSerializer.cpp` do
`build_src_filter`. Merge ho přidal jen do `[env:native]` a ASAN env pak padal
na undefined reference — což vypadá jako nález sanitizeru a je to jen chybějící
překladová jednotka. Ty dva envy musí zůstat v zákrytu.

Ověřeno: `pio test -e native` i `-e native_asan` 53/53, a čtyři envy, které
skutečně flashujeme, se přeloží (Xiao repeater / companion_radio_usb /
kiss_modem, t1000e_companion_radio_ble).

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-02 02:19:04 +02:00
tthandClaude Opus 5 c6fbd2fd2a fix(mesh): TRACE si připojoval cestu za konec payloadu
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-01 20:12:21 +02:00
tthandClaude Opus 5 8ac0da5bb4 fix(packet): readFrom četl pole dřív, než se zeptal, jestli tam vůbec jsou
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-01 20:12:01 +02:00
tthandClaude Opus 5 99981ff69b fix(mesh): markSeen je zase idempotentní, jinak si send path ujídá dedup okno
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-01 19:34:32 +02:00
tthandClaude Opus 5 ce9270ff21 fix(radio): fallback v calcMaxPacketMillis podtékal a uzel z toho ohluchl
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-01 19:34:16 +02:00
tthandClaude Opus 5 dc41906d72 feat(telemetry): dotaz na polohu nahřeje duty-cyklovanou GPS na příště
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-01 00:19:09 +02:00
tthandClaude Opus 5 6413230175 feat(t1000-e): gps duty cycle — deska ho neměla vůbec
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-31 22:54:07 +02:00
Liam Cottle 5edc936301 Merge pull request #3298 from meshcore-dev/companion-cli
Companion CLI
2026-08-27 18:17:10 +12:00
Scott Powell 55855f4cae * DynamicConfigSerializer fixes, and unit tests 2026-08-24 19:42:55 +10:00
Scott Powell 71deb27b7e * prefs, custom dirty state 2026-08-24 17:22:15 +10:00
Scott Powell 5a7bc507d1 * new DynamicConfigSerializer
* board/variant KeyValueStore now can write to 'custom' object in Json prefs
2026-08-24 17:12:56 +10:00
Scott Powell 8c73c948b4 * T096, and Station G3: refactoring the 'FEM' prefs to variant-specific code
* CommonCLI: 'FEM' commands removed
* introduced static no-op attachDynamicPrefs() for all boards
2026-08-24 01:50:22 +10:00
Scott Powell ae61e1a61c support for CMD_SEND_TXT_MSG and new TXT_TYPE_CLI_COMMAND 2026-08-23 22:49:45 +10:00
Scott Powell 644e68d0ff * new TXT_TYPE_CLI_COMMAND (3) 2026-08-23 22:04:05 +10:00
Scott Powell f66ed7ca88 * CommonRadioPrefs refactors done 2026-08-23 17:30:30 +10:00
Scott Powell 414e91fd5d * adding new CommonRadioPrefs
* refactor: moving various radio CLI handling to CommonRadioPrefs
2026-08-23 14:45:15 +10:00
liamcottle 1266fc058f don't toggle io pins on rak3401 when probing gps 2026-08-23 14:14:45 +12:00
tthandClaude Opus 5 acd235656f fix(mesh): šířku path hashe bere každá TX cesta z prefs uzlu, ne z konstanty
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-22 18:13:02 +02:00
tthandClaude Opus 5 18451dd559 fix(mesh): sendZeroHop mazal šířku path hashe, uzel se hlásil jako 1bajtový
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-22 15:04:52 +02:00
Scott Powell 5eaf5a86fb * initial wiring/routing of CLI commands to companion (either via app/serial interface, or remotely via txt message) 2026-08-21 20:55:57 +10:00
ripplebiz 1a8d860d8d Merge pull request #3233 from entr0p1/fix/centralise-max-clients
Fix duplicate MAX_CLIENTS definition
2026-08-20 14:09:30 +10:00
ripplebiz acc68d4fd4 Merge pull request #3234 from agessaman/fix/sensors-bosch-alt-i2c-address
Fix I2C address claiming and probe both BMP/BME addresses
2026-08-20 14:01:49 +10:00
Thomas Osterried cedc9977c2 Skip empty channel slots in searchChannelsByHash
An unconfigured slot has an all-zero secret, so it matches null-key group traffic (a sender with an unset PSK). The zero-key MAC validates against the empty slot and the foreign message is delivered as if it belonged to that channel — every node with a free slot is a null-key sink. Skip empty slots.
2026-08-19 09:35:25 +02:00
ripplebiz 6d8f16fabb Merge pull request #3246 from liamcottle/update/toolchain
Update GCC toolchain for nRF52 boards to fix builds on Raspberry Pi
2026-08-19 11:30:08 +10:00
liamcottle deb5e735b8 updated gcc toolchain for nrf52 boards 2026-08-19 12:07:29 +12:00
fdlamotte ed227b9cb6 Merge pull request #3180 from fdlamotte/thinknode_m8
thinknode_m8: initial support

adds some functionalities to ui_task and epddisplay:

epd display, 
- supports screen off, and clear screen
- supports forcing full refresh every N partial refreshes (off by default, configured to 60 for M8)

ui_task
- UI_SHOW_CLOCK flag to display clock on first screen
- rotary button can act like joystick (with regards to Enter key), defaults to old config for existing nodes
- screen off on long press when using rotary button (very handy for the watch)
2026-08-18 09:33:42 -04:00
Rastislav Vysoky 053fa3eb83 fix: move MeshadventurerBoard.h to correct location 2026-08-18 13:51:55 +02:00
agessaman b4917d6027 fix(sensors): probe BMP/BME at both 0x76 and 0x77
Grove and other Bosch modules strap SDO high, so the 0x76-only table
never initialized them. Add an alternate-address entry per Bosch sensor;
the bus scan still gates every probe, and all four drivers verify a chip
ID before claiming an address.

Each sensor type has a single static driver instance, so skip an entry
whose query is already active: the alternate address is a fallback, not
a second device.
2026-08-17 14:59:59 -07:00
agessaman 1a82394d72 fix(sensors): claim an I2C address after a successful init
Several table entries share an address and not every driver verifies a
chip ID: INA226::begin() only checks that the address ACKs, so an SHT4x
at 0x44 was also registered as an INA226 and reported junk current on a
second channel. Mark the address consumed once a driver initializes it
so later entries cannot re-claim the same device.
2026-08-17 14:58:49 -07:00
Florent cec44c05dd gxepd: do full refresh before screen off ... so wakes quicker ;) 2026-08-17 09:41:51 -04:00
Huw Duddy debbbb81c0 trimmed comment 2026-08-17 23:16:01 +10:00
entr0p1 11a02eeaba Duplicate MAX_CLIENTS definition
MAX_CLIENTS is defined in both src/helpers/ClientACL.h and examples/simple_repeater/MyMesh.h. Appears it was centralised in the former header file some time ago. Both are included in some places and, depending on which order they're in, either value can win.

This change drops the duplicate entry from the repeater firmware and bumps the central limit up to 32 (per the original repeater firmware value).

Changes:
- Remove MAX_CLIENTS from repeater MyMesh.h
- Increase MAX_CLIENTS limit in ClientACL.h to 32
2026-08-17 23:01:52 +10:00
Florent 3f4be26003 gxepddisplay: better handling of full refresh (no faint partial afterwards !!!) 2026-08-16 11:40:13 -04:00
tthandClaude Opus 5 cb74cdee06 Merge upstream v1.17.1 (origin/main)
Small one after v1.17.0: 57 files, +825/-200, and it merged with no conflicts at
all. Unlike 1.17.0, the release tags sit directly on origin/main, so no
`-s ours` is needed for git describe to see the release.

What matters for our nodes:

* fix-rxgain-setting-post-agc-reset -- rx gain no longer resets after an AGC
  adjustment. The release notes ask repeater admins to verify radio.rxgain (and
  the new radio.fem.rxgain) after updating, so that goes on the post-flash list.
* NRF52 CC310 crypto is initialised once instead of on every call, and several
  NRF52 pin maps were corrected.
* New 'get/set radio.fem.rxgain/txgain' CLI commands (companion cannot configure
  them yet).

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-08-15 14:13:50 +02:00
Fedor Kallay aa2292459f fix(lr2021): setTxPower() could leave the receiver down until reboot
RadioLibWrapper::setTxPower() called setOutputPower() and nothing else. On
LR2021 that writes the PA config and TxParams, which are standby-only commands,
and it never re-arms the receiver.

This only affects LR2021 boards. recvRaw() has:

    #if defined(USE_LR2021)
      state = STATE_RX;    // LR2021 stays in Rx after readData
    #else
      state = STATE_IDLE;  // need another startReceive()
    #endif

so on SX126x the next recvRaw() re-arms Rx anyway and the write is harmless,
while on LR2021 startReceive() is never called again on its own. A 'set tx'
issued while the radio was listening could therefore stop reception until the
next reboot. Observed a few times on an LR2021 repeater; recovery required a
power cycle.

Dispatcher's stuck-radio check does not catch it: isInRecvMode() reports the
wrapper's own `state` flag rather than the chip's actual mode, so it still
believes the radio is in Rx, and it only raises ERR_EVENT_STARTRX_TIMEOUT
without attempting recovery.

Fix: on LR2021, drop to standby via idle() before writing the PA config and let
checkRecv() re-arm Rx - the same pattern resetAGC() and
applySideDetectorConfig() already use. Other radios are left untouched.

Affects meshtracker_x1 and meshnology_w12 (both USE_LR2021).

Tested on hardware with an LR2021 repeater: six consecutive 'set tx' changes
(15/18/21/14/19/14), after which the radio kept receiving and forwarding
traffic (rawrx/rxpkts increasing, no missed IRQs). Noise-floor sampling also
kept updating, which only happens while state == STATE_RX.
2026-08-15 02:31:19 +02:00
Scott Powell 10b6552872 * combine radio's entropy with CC310 RNG 2026-08-14 17:46:43 +10:00
ripplebiz f8832af086 Merge pull request #3106 from ViezeVingertjes/fix/scoped-reply-routing
Fix replies dropped when flood.max.unscoped is low
2026-08-13 15:00:37 +10:00
ripplebiz 09933264fb Merge pull request #3188 from oltaco/heltec-t096-pin-fix
Fix Heltec T096 Tx issue
2026-08-12 23:00:31 +10:00
taco 0f686aa3f1 Revert "t096 compiler layout workaround"
This reverts commit bdefc8d085.
2026-08-12 22:28:36 +10:00
ripplebiz d5949f93d8 Merge pull request #3137 from agessaman/feat/station-g3-fem-prefs
Station G3: Expose, persist, and apply FEM gain preferences
2026-08-12 16:11:23 +10:00
ripplebiz c913b0fec1 Merge pull request #3182 from oltaco/t096-tx-workaround
Workaround for Heltec T096 transmit issue
2026-08-12 14:49:32 +10:00
taco bdefc8d085 t096 compiler layout workaround
see https://github.com/meshcore-dev/MeshCore/issues/3151
2026-08-12 14:04:50 +10:00
ripplebiz 3dfbcf93b8 Merge pull request #3154 from fdlamotte/nrf52_hwcrypto_init
nrf52: call nRFCrypto.begin()/end() only once
2026-08-12 12:14:21 +10:00
Florent 109ed685b2 thinknode_m8: initial support 2026-08-11 17:37:35 -04:00
ripplebiz c6d2dd6234 Merge pull request #3164 from weebl2000/fix-tbeam-s3-display
Fix display issues on T-beam Supreme S3
2026-08-11 14:39:39 +10:00
agessaman f82950f311 feat(companion_radio): add external FEM gain preferences for RX and TX for companions
Introduced consistent preferences for external LoRa FEM RX and TX gain settings in NodePrefs. Updated companion MyMesh to apply these settings during initialization and transmission. Added unit tests to verify the round-trip serialization of these new preferences.
2026-08-10 11:59:26 -07:00