PP3K Video Player tight loop

  asm volatile(
    "ldi    r16,1\n\t"
    "mov    __tmp_reg__,r16\n\t"
    "clr    __zero_reg__\n\t"
    "clr    r22\n\t"
    "clr    r23\n\t"
    "clr    r24\n\t"
    "clr    r25\n\t"
    "ldi    r16, 0xff\n\t"

    // wait for datablock start (0xfe)
  "waitForDBStart:\n\t"
    "in     r17, 0x2d\n\t"
    "in     r17, 0x2e\n\t"
    "out    0x2e, r16\n\t"
  "waitForDBStartResponse:\n\t"
    "in     r17, 0x2d\n\t"
    "sbrs   r17, 7\n\t"
    "rjmp   waitForDBStartResponse\n\t"
    "in     r17, 0x2e\n\t"
    "cpi    r17, 0xfe\n\t"
    "brne   waitForDBStart\n\t"

    "clr    r20\n\t"
    "clr    r21\n\t"

    // initiate first SPI transfer of data block
    "out    0x2e, r16\n\t"
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through
    "nop\n\t"  // wait for the first byte to come through

  "SPItoScreenLoop:\n\t"
    // Read last byte from SPI
    "in     r17, 0x2e\n\t"
    // initiate next SPI transfer while we output to screen
    "out    0x2e, r16\n\t"

    // low nibble to PORTF high
    "mov    r19, r17\n\t"
    "lsl    r19\n\t"
    "lsl    r19\n\t"
    "lsl    r19\n\t"
    "lsl    r19\n\t"
    "out    0x11, r19\n\t"
    // high nibble to PORTD low
    "in     r18, 0x0b\n\t"
    "andi   r18, 0xf0\n\t"
    "lsr    r17\n\t"
    "lsr    r17\n\t"
    "lsr    r17\n\t"
    "lsr    r17\n\t"
    "or     r18, r17\n\t"
    "out    0x0b, r18\n\t"
    // Toggle ILI WR
    "cbi    0x08, 6\n\t"
    "sbi    0x08, 6\n\t"

    // Check if we are done
    "add    r20, __tmp_reg__\n\t"
    "adc    r21, __zero_reg__\n\t"
    "cpi    r21,2\n\t" // test against 512
    "brne   jmpToStart\n\t"
    "cpi    r20,0\n\t" // test against 513 as we skip the first write to let us write to screen while waiting for SPI
    "brne   jmpToStart\n\t"

    "rjmp   readCRC\n\t"

  "jmpToStart:\n\t"
    "rjmp   SPItoScreenLoop\n\t"

  "readCRC:\n\t"
    "in     r17, 0x2d\n\t"
    "in     r17, 0x2e\n\t"
    "out    0x2e, r16\n\t"
  "readCRCResponse:\n\t"
    "in     r17, 0x2d\n\t"
    "sbrs   r17, 7\n\t"
    "rjmp   readCRCResponse\n\t"

  "checkLoop:\n\t"
    "add    r22, __tmp_reg__\n\t"
    "adc    r23, __zero_reg__\n\t"
    "adc    r24, __zero_reg__\n\t"
    "adc    r25, __zero_reg__\n\t"
    "cpi    r22, 0xd6\n\t"
    "brne   jumpToDBStart\n\t"
    "cpi    r23, 0x07\n\t"
    "brne   jumpToDBStart\n\t"
    "cpi    r24, 0x04\n\t"
    "brne   jumpToDBStart\n\t"
    "cpi    r25, 0x00\n\t"
    "brne   jumpToDBStart\n\t"
    "rjmp   end\n\t"

  "jumpToDBStart:\n\t"
    "rjmp   waitForDBStart\n\t"

  "end:\n\t"
    :
    :
    : "r16", "r17", "r18", "r19", "r20", "r21", "r22", "r23", "r24", "r25"
  );