// csamul8.e  four csa4 used to make 8 x 8 bit multiplier
//            uses add8c.e and csa4.e components

// a times b = p, product

define csamul8(a[8], b[8], p[16])
  signal z4[4] <= #b0000;
  signal z3[3] <= #b000;
  signal pt[3]; // partial product
  signal ps[3];
  signal co[8]; // partial carry
  signal c1[8];
  signal c2[8];
  signal cout;  // unused output of adder
  signal so[8]; // partial sum
  signal s1[8];
  signal s2[8];
  signal s3[8];
circuits
  mul0 use csa4(a[3:0], b[3:0], z4, z4, pt, so[3:0], co[3:0], p[2:0]);
  mul1 use csa4(a[7:4], b[3:0], z4, z4, z3, so[7:4], co[7:4], pt);
  mul2 use csa4(a[3:0], b[7:4], s1[3:0], co[3:0], ps, s2[3:0], c1[3:0], p[6:4]);
  mul3 use csa4(a[7:4], b[7:4], s1[7:4], co[7:4], z3, s2[7:4], c1[7:4], ps);
  add use add8c(s3, c1, #b0, p[15:8], cout);
  s1   <= #b0.so[7:1] after 1ns;
  p[3] <= so[0] after 1ns;
  s3   <= #b0.s2[7:1] after 1ns;
  p[7] <= s2[0] after 1ns;
end circuits;
end csamul8;