Embedded Systems February 10, 2016

Serial Interface - SPI  Serial Peripheral Interface  Synchronous communications  Clock supplied by the Master (The unit controlling the communications)  Full duplex communications  Interface with devices like:  LCD displays  Sensors  Cameras  Real Time Clocks  Memory

SPI  Interfacing with LCD module  Left side (Master) would be the MCU  Right side (Slave) would be the LCD  Each clock cycle, shifts one bit into the Slave  Clock speed determined by the master  SS = Slave Select

SPI configuration  SPI Control Register 1 (SPICR1)  SPI Control Register 2 (SPICR2)  SPI Baud Rate Register (SPIBR)  SPI Status Register (SPISR)  SPI Data Register (SPIDR)

Serial Interface  Serial Communications Interface  Asynchronous communications  Each device has its own serial clock  Baud rate negotiated before communications  Interface with devices like:  LCD displays  Sensors  Computers  Printers

SCI Data Format

SCI Interface configuration  SCI Baud Rate Registers (SCIBDH and SCHBDL)  SCI Control Register 1 (SCICR1)  SCI Control Register 2 (SCICR2)  SCI Status Register 1 (SCISR1)  SCI Status Register 2 (SCISR2)  SCI Data Registers (SCIDRH and SCIDRL)

Synchronization between devices  Software Latency: The time between when the I/O device needs service, and the time when service is initiated.  For an input device, latency is the time between new input data ready, and the software reading the data.  For an output device, latency is the time between output device idle, and the software giving the device new data to output.  Periodic events (sampling ADC, outputting to DAC). Latency is the time between - when it is supposed to be run, and when it is actually run. taken from notes by: Jonathan W. Valvano

Synchronization between devices  A real time system is one that can guarantee a worst case software latency. In other words, there is an upper bound on the software response time.  Hardware Latency or device latency is the time between when an I/O device is given a command, and the time when command is completed. taken from notes by: Jonathan W. Valvano

Synchronization between devices  I/O bound  Bandwidth is limited by speed of I/O device  Making the I/O device faster will increase bandwidth  Making the software run faster will not increase bandwidth  Software often waits for the I/O device  CPU bound  Bandwidth is limited by speed of executing software  Making the I/O device faster will not increase bandwidth  Making the software run faster will increase bandwidth  Software does not have to wait for the I/O device taken from notes by: Jonathan W. Valvano

Synchronization  Five ways to synchronize  Blind Cycle counting  Busy Waiting  Interrupt  Periodic Polling  Direct Memory Access taken from notes by: Jonathan W. Valvano

Blind Cycle Counting taken from notes by: Jonathan W. Valvano

Gadfly or Busy Waiting taken from notes by: Jonathan W. Valvano

Interrupts  When data is read or written, the processor is interrupted from current task  We will study interrupts in details so let’s leave it for now

Periodic Polling  Write value  Check if value is processed  Do other stuff  Come back and check if value is processed

Direct Memory Access  I/O device reads/writes directly to RAM  We will not worry about this synchronization in this course taken from notes by: Jonathan W. Valvano