; TMS320C54x assembly language code
; compute a polynomial y(x) = c0 + c1 x + c2 x^2 + c3 x^3
; using Horner's form  y(x) = c0 + x (c1 + x (c2 + x (c3)))
; ar2 contains address of array [c3 c2 c1 c0]
; poly uses temporary register t to store multiplicand x
; numbers are assumed to be in fixed-point format in [-1, 1)
; four iterations of execution of the poly instruction with
; the value of b being intially c3 << 16
;   1. a = c3 + x * 0 = c3;                  b = c2
;   2. a = c2 + x * c3;                      b = c1
;   3. a = c1 + x * (c2 + x * c3);           b = c0
;   4. a = c0 + x (c1 + x * (c2 + x * c3));  b = c?
; Programmer: Brian L. Evans, UT Austin
	ld   *ar2+,16,b	; b = c3 << 16
	ld   *ar3,t	; t = x (ar3 contains address of x)
	rptz a,#3	; a = 0, repeat next instruction 4 times
	poly *ar2+	; a = b + x * a || b = c(i-1) << 16
	sth  a,*ar4	; store result (ar4 contains address of y)