double cfir1(int M, double* h,
             double* w, double** p, double x) {

*(*p)-- = x;
wrap(M, w, p);

   for (double y=0, h+=M, int i=M; i>=0; i--) {
      y += (*h--) * (*(*p)--);
      wrap(M,w,p);
   }
   return y;
}

NEWX     LDF    IN, R3                read new input sample
         STF    R3, *AR1++%           put x in a register and increment
         LDF    0.0, R0               initialize accumulators
         LDF    0.0, R2

         RPTS   M                     repeat for i=M to i=0
         MPYF3  *AR0++%, *AR1++%, R0  calculate and in parallel
     ||  ADDF3  R0, R2, R2            accumulate

         ADDF   R0, R2                accumulate last product
         STF    R2, Y                 output
         B      NEWX                  branch and repeat