#	PROGRAM TO COMPUTE FAXPY
#	P0:mux.2.brcst	R3, R3		#copy scalar "a" to all 16-bit subwords in R3
#foo:
 	P1:LOAD.8.upd	R4, 8(R1)	#load (X(i), X(i+1), X(i+2), X(i+3)), update R1
	P1:PMUL.odd	R5, R4,R3	#multiply (a*X(i), a*X(i+2))
	P1:PMUL.even	R6, R4,R3	#multiply (a*X(i+1), a*X(i+3))
	P1:LOAD.8	R8, 8(R2)       #load (Y(i), Y(i+1), Y(i+2), Y(i+3))
	P1:MIX.2.L	R7, R8,R0	#expand 16-bit to 32-bit subwords, padding on right
	P1:MIX.2.R	R8, R8,R0	#R7=(Y(i), Y(i+2)) and R8=(Y(i+1), Y(i+3))
	P1:PADD.4	R9, R5,R7       #add (a*X(i)+Y(i), a*X(i+2)+Y(i+2))
	P1:PADD.4	R10, R6,R8      #add (a*X(i+1)+Y(i+1), a*X(i+3)+Y(i+3))
	P1:MIX.2.L	R8, R9,R10	#contract 32-bit down to 16-bit subwords
	P1:STORE.8.upd	8(R2), R8	#store new (Y(i), Y(i+1), Y(i+2), Y(i+3)), update R2
	P1:CMP.leu	R1,R11, P1,P0	#test if done (result to P0 is discarded)
	P1:JMP		-44		#loop if not done