«

I have a really simple setup to test the initialization of signals, and
am a little confused as to the simulator results.

When the compiled design is loaded into the simulator (ModelSim), sig2
is initialized at 'U' and sig3 is initialized at '1'.

sig2 is initialized in the signal declaration part of the top-level
file, whereas sig3 is initialized in the entity port. Why is there a
difference in the simulator? Here is the code:

CLIENT:

LIBRARY ieee;
USE ieee.std_logic_1164.ALL;
USE ieee.std_logic_arith.ALL;
USE ieee.std_logic_unsigned.ALL;

ENTITY client IS
PORT(
sig2 : OUT std_logic;
sig3 : OUT std_logic := '1'
);
END client ;
--
ARCHITECTURE struct OF client IS
BEGIN
END struct;




SERVER:

LIBRARY ieee;
USE ieee.std_logic_1164.ALL;
USE ieee.std_logic_arith.ALL;
USE ieee.std_logic_unsigned.ALL;

ENTITY server IS
PORT(
sig2 : IN std_logic;
sig3 : IN std_logic
);
END server ;
--
ARCHITECTURE struct OF server IS
BEGIN
END struct;





TOP-LEVEL:

LIBRARY ieee;
USE ieee.std_logic_1164.ALL;
USE ieee.std_logic_arith.ALL;
USE ieee.std_logic_unsigned.ALL;

ENTITY test_top IS
END test_top ;

LIBRARY test;

ARCHITECTURE struct OF test_top IS
-- Internal signal declarations
SIGNAL sig2 : std_logic := '1';
SIGNAL sig3 : std_logic;

-- Component Declarations
COMPONENT client
PORT (
sig2 : OUT std_logic ;
sig3 : OUT std_logic := '1'
);
END COMPONENT;
COMPONENT server
PORT (
sig2 : IN std_logic ;
sig3 : IN std_logic
);
END COMPONENT;

BEGIN
-- Instance port mappings.
I0 : client
PORT MAP (
sig2 => sig2,
sig3 => sig3
);
I1 : server
PORT MAP (
sig2 => sig2,
sig3 => sig3
);

END struct;

Your sig2 is an output from client model and hence that will win over
the local signal initialization done at top_level. Since sig2 is not
being driven by your client model, it stays at U.


Sig3 is initialized by the client model's output itself.

Does that explain?

Regards
Ajeetha, CVC
www.noveldv.com
* A Pragmatic Approach to VMM Adoption 2006 ISBN 0-9705394-9-5
h**p://www.systemverilog.us/
* SystemVerilog Assertions Handbook
* Using PSL/Sugar


Analog_Guy wrote:
> I have a really simple setup to test the initialization of signals, and
> am a little confused as to the simulator results.
>