Home › Functional Coverage › Transition Coverage

Transition Coverage

Statical values are good, but verifying sequences (A followed by B) is better. Transition coverage allows you to track value changes over consecutive sample events.

Basic Transitions (=>)

Use the => operator inside a bin definition to specify a sequence. 

covergroup cg_fsm @(posedge clk);
    cp_state: coverpoint state {
        // Did we go from IDLE (0) directly to BUSY (1)?
        bins idle_to_busy = (0 => 1);
        
        // Did we go from BUSY (1) back to IDLE (0)?
        bins busy_to_idle = (1 => 0);
    }
endgroup

Complex Sequences

You can chain multiple steps and use ranges. 

cp_sequence: coverpoint data {
    // 3-step sequence: 0 -> 1 -> 2
    bins count_up = (0 => 1 => 2);

    // Any value from 0-3 followed by any value from 4-7
    bins low_to_high = ([0:3] => [4:7]);
    
    // Repetition: 1, followed by 1 (at least twice)
    bins stable_high = (1 [* 2]); 
    
    // Goto: 1, then anything, then 3 (non-consecutive)
    bins one_eventually_three = (1 => ... => 3); 
}

Note

Transition coverage is strictly synchronous to the sampling event of the covergroup.

Common Interview Questions

Q: Can I use wildcards in transition bins?
No, you cannot explicitly use a wildcard character like `*` inside the transition sequence in the same way you might expect. However, you can list multiple values or ranges locally, e.g., `(0,1,2 => 3)`.
Q: What is the difference between (0=>1) and (0,1)?
(0 => 1) is a temporal transition: Value 0 at time T, then Value 1 at time T+1. (0, 1) is a set: Value 0 OR Value 1 at any time T.
Previous Cross Coverage Next Coverage Options