Good evening,
I designed a structural design of the MC14585B magnitude comparator: https://www.onsemi.com/pub/Collateral/MC14585B-D.PDF
I wanted to simulate all 2^8 possibilities and so I wrote a testbench to just that. I am using modelsim student edition:
source:
module MC14585(
input [3:0] A, B,
input A_less_B_in, A_greater_B_in, A_equal_B_in,
output A_less_B_out, A_greater_B_out, A_equal_B_out);
wire [3:0] exor, not_A, not_B, nand1, or1;
wire not_A_less_B_in, not_A_equal_B_in, nand2, nor1;
not n1(not_A_less_B_in, A_less_B_in);
not n2(not_A_equal_B_in, A_equal_B_in);
xor x1(exor[0], A[0], B[0]);
xor x2(exor[1], A[1], B[1]);
xor x3(exor[2], A[2], B[2]);
xor x4(exor[3], A[3], B[3]);
not n3(not_A[0], A[0]);
not n4(not_A[1], A[1]);
not n5(not_A[2], A[2]);
not n6(not_A[3], A[3]);
not n7(not_B[0], B[0]);
not n8(not_B[1], B[1]);
not n9(not_B[2], B[2]);
not n10(not_B[3], B[3]);
nand na1(nand1[0], not_A[0], B[0]);
nand na2(nand1[1], not_A[1], B[1]);
nand na3(nand1[2], not_A[2], B[2]);
nand na4(nand1[3], not_A[3], B[3]);
or o1(or1[0], exor[3], exor[2], exor[1], exor[0], not_A_less_B_in);
or o2(or1[1], exor[3], exor[2], exor[1], nand1[0]);
or o3(or1[2], exor[3], exor[2], nand1[1]);
or o4(or1[3], exor[3], nand1[2]);
nand na5(nand2, or1[0], or1[1], or1[2], or1[3], nand1[3]);
nand na6(A_less_B_out, or1[0], or1[1], or1[2], or1[3], nand1[3]);
nor not_or1(nor1, exor[0], exor[1], exor[2], exor[3], not_A_equal_B_in);
nor not_or2(A_greater_B_out, nand2, nor1);
endmodule
testbench:
`timescale 1ns / 1ps
module Assignment1_tb();
reg[3:0] A, B;
wire A_less_B, A_greater_B, A_equal_B;
MC14585 MC14585_DUT(
.A(A),
.B(B),
.A_less_B_in(1'b0),
.A_greater_B_in(1'b0),
.A_equal_B_in(1'b1),
.A_less_B_out(A_less_B),
.A_greater_B_out(A_greater_B),
.A_equal_B_out(A_equal_B));
initial begin
for (A = 0; A < 16; A=A+1) begin
for (B = 0; B < 16; B=B+1) begin
if (A_less_B && (A < B))
$display ("%d is less than %d", A, B);
else if (A_greater_B && (A > B))
$display ("%d is greater than %d", A, B);
else if (A_equal_B && (A == B))
$display ("%d is equal to %d", A, B);
else
$display ("ERROR");
#10;
end
end
end
endmodule
For some reason when I run my testbench, the outer for loop (A) doesn't iterate values at all. Also, I placed this loop into an initial statement so it will only iterate all the way through once, but it continues to run until I end the simulation.
Here is some sample output:
# 0 is less than 2
# 0 is less than 3
# 0 is less than 4
# 0 is less than 5
# 0 is less than 6
# 0 is less than 7
# 0 is less than 8
# 0 is less than 9
# 0 is less than 10
# 0 is less than 11
# 0 is less than 12
# 0 is less than 13
# 0 is less than 14
# 0 is less than 15
# ERROR
# ERROR
# 0 is less than 2
# 0 is less than 3
# 0 is less than 4
# 0 is less than 5
# 0 is less than 6
# 0 is less than 7
# 0 is less than 8
# 0 is less than 9
# 0 is less than 10
# 0 is less than 11
# 0 is less than 12
# 0 is less than 13
# 0 is less than 14
# 0 is less than 15
# ERROR
# ERROR
# 0 is less than 2
# 0 is less than 3
# 0 is less than 4
# 0 is less than 5
# 0 is less than 6
# 0 is less than 7
# 0 is less than 8
# 0 is less than 9
any ideas what's going on? Thanks
Your problem is in these statements:
reg[3:0] A, B;
...
for (B = 0; B < 16; B = B + 1)
since B is '4' bits wide, it will never be greater or equal to 16. When it gets 15 (4'b1111), the next increment will overflow and make it '0'.
You need make B wider than 4 bits.