TODO.txt

* model: ADC should be an interface, not a device
* model: see if SPI slaves really need to be threaded (just rely on the SPI lock ?)
* model: MAX111x::Channel
* model: add new class Queue<type> (message queues)
* model: implement a full init chain (no ref. count)

* feature: A25Lxxx low power mode
* feature: IRQ/VIC registers
* feature: Device::reset()
* feature: FlashArray: address mask ( sn < _nSlaves )
* feature: FlashMemory: mask page/sector address
* feature: enable() / disable() should lock the device before calling incRef() / decRef()
    * should all devices be threaded ?
    * would it be enough to just suspend the scheduler during incRef() / decRef() ?
      -> in that case, could an interrupt cause a context switch anyway, while the scheduler is suspended ?
      -> see FreeRTOS sources / ask the FreeRTOS guys

* misc: find and hunt down C standard library headers


DONE

* 20151020 - model: SPIStream -> SPI::Stream
* 20151020 - model: DMAStream -> DMA::Stream
* 20151020 - model: GPIOPin   -> GPIO::Pin
* 20151019 - model: BusDevice -> BusSlave


20151105

. test reset hard + VBAT (latchup)
. sortir la version du soft sur le port debug
. tester la (vraie) tempo 30mn
. calibrer les ADC (voltage, convertisseur)
. parametre de puissance CW
. parametres des commandes
. valeur par defaut CW
. indicatif 1 fois sur 10
. toujours du CW en mode powersave ?!
. sauvegarde mode precedant pour la sortie du powersave (donc non, on reste en WODEX)