* polyphase.c - final stage of subband transform (polyphase synthesis filter)
* Look in the appropriate subdirectories for optimized asm implementations
*   (e.g. arm/asmpoly.s)

Multiply with 64-bit result    32    UMULL, SMULL, UMLAL, and SMLAL.
unsigned accumulate long UMLAL{S} RdLo, RdHi, Rm, Rs N Z C* V* RdHi,RdLo := unsigned(RdHi,RdLo + Rm * Rs)

void decode_start(void)
{
   d_start = RIT->RICOUNTER;
}

void decode_end(void)
{
   d_tot += RIT->RICOUNTER - d_start;
}

void play_start(void)
{
   //reset tottal decode time counter
   d_tot = 0;
   //reset
   RIT->RICOUNTER = 0;
   //start
   RIT->RICTRL = (1 << 0) |  //clear int flag
                 (0 << 1) |  //don't clear timer on equal
                 (1 << 2) |  //debug mode - halted on cpu halt
                 (1 << 3) |  //enable
                 (0);
}

void play_stop(void)
{
   unsigned int tot;
   //
   tot = RIT->RICOUNTER;
   //
   RIT->RICTRL &= ~(1 << 3); //stop counter
   //
   printf("Total ticks  = %d\r\n", tot);
   printf("Decode ticks = %d\r\n", d_tot);
   printf("Average load = %d %%\r\n", ((long long)d_tot * 100) / (long long)tot);
}

    decode_start();
     err = MP3Decode(hMP3Decoder, &read_ptr, &bytes_left, dac_buffer[writeable_buffer], 0);
     decode_end();