The "relax" option enables the linker to convert certain "call"
instructions to the smaller "rcall" instruction.  This reduces the
size of the resulting binary.

The big LCD code update. This splits the display and menu implementation. Paving the way for future different displays. It also makes it easier to modify the menu structures while keeping everything consistant. Note that this most likely breaks the translations, which need to be updated.

Add PID functional range setting. With high powered heaters the current functional range of 10 degrees causes high overshoots as the PID needs to kick in before the temperature hits 10 degrees below target.

Fix the sensitive pin definitions, there where analogue numbers in the digital pin list. Also made M42 without a P function on the LED_PIN (which was otherwise a useless pin definition)

Fix endstop pins for Printrboard

Added some missing Mxxx comments to M commands. Fixed the unsetting of the timer dividers for the fast PWM fan.

Move the Arduino related stuff out of the Marlin tree, as they are not directly related to Marlin functionality but addons for the Arduino IDE. Also split up the addons for pre 1.0.0 Arduino and post 1.0.0 Arduino.

Use the disable macros on initialization. Which fixes a bug when there is no pin defined for stepper enabled (motherboard 4)

Add Printrboard and Brainwave to pins.h and fastio.h

Big temperature code update. No longer converts back and forwards between temperature and raw sample value. Reducing complexity, removing code. Also named some variables better. While keeping the safety intact and functionality the same.

Stop redefining MOTHERBOARD and #define AT90USB instead. Also put #ifndef AT90USB around the HardwareSerial_h trick, so now Printrboard / Brainwave support compiles on Arduino 1.0.2 as well.

Add support for Printrboard (from https://github.com/lincomatic/Marlin) and Brainwave (from https://github.com/metrix/Sprinter).

Small makefile change, posibility to add defines on the commandline, and fix compiling with watchdog enabled when using the Makefile.

Fix the WATCHPERIOD implementation. It did not work correctly in a multi extruder setup, it did not work after 32 seconds after startup (int16 used as millies timestamp). And it did not work if you gave an M104 or M109 when the target was already around the target setpoint. So on average, it did not work at all. The new implementation should be robust in detecting a failure to heat up.