Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

shift_register_compare assert question

Status
Not open for further replies.

promach

Advanced Member level 4
Joined
Feb 22, 2016
Messages
1,199
Helped
2
Reputation
4
Reaction score
5
Trophy points
1,318
Activity points
11,636
Why would temporal induction fail in less than 32 clock steps ? induction waveform

I suppose 16 clock steps would be sufficient since after 16 clock steps, both shift registers sa and sb will be equivalent no matter how different they are at clock step #0

shift_register_compare assert question.png


Code Verilog - [expand]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
/*
 assume(i_ce) requires 16 steps, 
 always @(posedge i_clk) if (!$past(i_ce)) assume(i_ce); should take 16*2, 
 always @(posedge i_clk) if ((!$past(i_ce))&&(!$past(i_ce,2))) assume(i_ce); should require 16 * 3 = 48 steps.
 
Here's what's going on: there are 16 values in that shift register.  It takes i_ce being high to flush one more stage through the assertion.  If assume(i_ce), then it will take about 16 clocks to flush the state in the shift register until sa == sb;                                                                                
 
if you do an always @(posedge i_ce) if (!$past(i_ce)) assume(i_ce); then there will never be more than one cycle between times when i_ce is high.                    
 
The induction engine is going to prove us wrong, so it will stretch out the distance between the i_ce's as far as it can.  If we require that it cannot stretch out by more than 2 samples, then it can only stretch out the time to prove all of the shift register elements to 32 clocks at the most (16 shift register elements, times maximum two clocks per i_ce)  
 
*/
 
module shift_register_compare(clk, i_ce, i_bit);
 
input clk;
input i_ce;
input i_bit;
 
parameter SHIFT_REG_LENGTH = 15;
 
reg[SHIFT_REG_LENGTH:0] sa;
reg[SHIFT_REG_LENGTH:0] sb;
 
initial sa = 0;
initial sb = 0;
 
always @(posedge clk)
    if(i_ce)
        sa <= { sa[(SHIFT_REG_LENGTH-1):0] , i_bit };
 
always @(posedge clk)
    if(i_ce)
        sb <= { sb[(SHIFT_REG_LENGTH-1):0] , i_bit };
 
always @(*)
    assert(sa[SHIFT_REG_LENGTH] == sb[SHIFT_REG_LENGTH]);
 
always @(posedge clk) 
    if (!$past(i_ce)) assume(i_ce); 
 
endmodule

 

There is something wrong with your testbench. You have a race condition. The simulation shown has two different values i sa and sb can't happen basex on your posted code.

Didn't you read my other post about shifting testbench signals that drive your dut away from the clock edge?
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top