由于串行多位加法器的高位相加時要等待低位的進(jìn)位，所以速度受到進(jìn)位信號的限制而變慢，人們又設(shè)計了一種多位數(shù)超前進(jìn)位加法器邏輯電路，使每位求和結(jié)果直接接受加數(shù)和被加數(shù)而不必等待地位進(jìn)位，而與低位的進(jìn)位信號無關(guān)，這就大大的提高了運算速度。現(xiàn)在簡單介紹超前進(jìn)位的運算方法，以及VHDL可編程邏輯編程。
Library ieee;
Use ieee.std_logic_1164.all;
entity CQadd_Co is
port(C0,A1,B1,A2,B2,A3,B3,A4,B4:In std_logic;
S1,S2,S3,S4,Co:out std_logic);
end CQadd_Co;
architecture adder of CQadd_Co is
signal c1,c2,c3,t1,t2,t3,t4,m1,m2,m3,m4:std_logic;
signal P1,P2,P3,P4:std_logic;
begin
t1<=not(A1 AND B1);m1<=not(A1 or B1);
t2<=not(A2 AND B2);m2<=not(A2 or B2);
t3<=not(A3 AND B3);m3<=not(A3 OR B3);
t4<=not(A4 AND B4);m4<=not(A4 OR B4);
P1<=(not m1)and t1;P2<=(not m2)and t2;
P3<=(not m3)and t3;P4<=(not m4)and t4;
c1<=not(((not C0)and t1)or m1);
c2<=not(((not C0)and t1 and t2)or (t2 and m1) or m2);
c3<=not(((not C0)and t1 and t2 and t3)or (t2 and t3 and m1) or (m2 and t3)or m3);
Co<=not(((not C0)and t1 and t2 and t3 and t4)or(m1 and t4 and t3 and t2)or(t2 and t3 and t4)or(t4 and m3)or m4);
S1<=P1 XOR C0;S2<=P2 XOR c1;S3<=P3 XOR c2;S4<=P4 XOR c3;
end;