[avr] Improved SPI speed on 16bit transfer.

From arduino/Arduino#2376 (comment)

Quoting Andrew Kroll:

   [..this commit..] introduces a small delay that can prevent the wait
   loop form iterating when running at the maximum speed. This gives
   you a little more speed, even if it seems counter-intuitive. At
   lower speeds, it is unnoticed. Watch the output on an oscilloscope
   when running full SPI speed, and you should see closer back-to-back
   writes.

Quoting Paul Stoffregen:

   I did quite a bit of experimenting with the NOP addition. The one
   that's in my copy gives about a 10% speedup on AVR.

Author: cmaglie

avr/libraries/SPI/SPI.h

@@ -184,6 +184,12 @@ class SPIClass {
   // Write to the SPI bus (MOSI pin) and also receive (MISO pin)
   inline static uint8_t transfer(uint8_t data) {
     SPDR = data;
+    /*
+     * The following NOP introduces a small delay that can prevent the wait
+     * loop form iterating when running at the maximum speed. This gives
+     * about 10% more speed, even if it seems counter-intuitive. At lower
+     * speeds it is unnoticed.
+     */
     asm volatile("nop");
     while (!(SPSR & _BV(SPIF))) ; // wait
     return SPDR;
@@ -193,16 +199,20 @@ class SPIClass {
     in.val = data;
     if (!(SPCR & _BV(DORD))) {
       SPDR = in.msb;
+      asm volatile("nop"); // See transfer(uint8_t) function
       while (!(SPSR & _BV(SPIF))) ;
       out.msb = SPDR;
       SPDR = in.lsb;
+      asm volatile("nop");
       while (!(SPSR & _BV(SPIF))) ;
       out.lsb = SPDR;
     } else {
       SPDR = in.lsb;
+      asm volatile("nop");
       while (!(SPSR & _BV(SPIF))) ;
       out.lsb = SPDR;
       SPDR = in.msb;
+      asm volatile("nop");
       while (!(SPSR & _BV(SPIF))) ;
       out.msb = SPDR;
     }