When to use non-blocking (sequential) in Verilog - Time & Space Complexity

Time Complexity: When to use non-blocking (sequential)

O(N)

Understanding Time Complexity

We want to understand how the time it takes to run sequential code with non-blocking assignments grows as the input size changes.

How does the number of steps change when we use non-blocking assignments in sequential logic?

Scenario Under Consideration

Analyze the time complexity of the following Verilog sequential block using non-blocking assignments.

always @(posedge clk) begin
  for (integer i = 0; i < N; i++) begin
    reg_array[i] <= input_array[i];
  end
end

This code copies values from one array to another on each clock cycle using non-blocking assignments.

Identify Repeating Operations

Look for loops or repeated actions inside the sequential block.

• Primary operation: The for-loop that copies each element from input_array to reg_array.
• How many times: The loop runs N times every clock cycle.

How Execution Grows With Input

Each clock cycle, the number of copy operations grows with N.

Input Size (N)	Approx. Operations per cycle
10	10 copies
100	100 copies
1000	1000 copies

Pattern observation: The work grows directly with the size of the array. Double the array size, double the operations.

Final Time Complexity

Time Complexity: O(N)

This means the time to complete the copying grows linearly with the number of elements.

Common Mistake

[X] Wrong: "Non-blocking assignments run all at once instantly, so time does not grow with N."

[OK] Correct: Even though non-blocking assignments update registers simultaneously at the clock edge, the hardware still needs to handle each element, so the total work grows with the number of elements.

Interview Connect

Understanding how sequential blocks with non-blocking assignments scale helps you design efficient hardware and explain your design choices clearly.

Self-Check

What if we replaced the for-loop with parallel assignments for each element? How would the time complexity change?

Practice

1. When should you use non-blocking assignments (<=) in Verilog?

easy

A. To declare constants in your design

B. To model sequential logic that updates on clock edges

C. To model combinational logic with immediate updates

D. To assign values outside of always blocks

Solution

1. Step 1: Understand the purpose of non-blocking assignments

   Non-blocking assignments (<=) are used to model sequential logic, which updates values on clock edges, like flip-flops.

2. Step 2: Differentiate from other assignment types

   Blocking assignments (=) are used for combinational logic with immediate updates, not sequential logic.

3. Final Answer:

   To model sequential logic that updates on clock edges -> Option B

4. Quick Check:

   Non-blocking = sequential logic [OK]

Hint: Use non-blocking for clocked (sequential) logic only [OK]

Common Mistakes:

• Using non-blocking for combinational logic
• Confusing blocking and non-blocking assignments
• Assigning outside always blocks

2. Which of the following is the correct syntax for a non-blocking assignment inside an always block triggered on a clock edge?

easy

A. always @(posedge clk) begin a <= b; end

B. always @(posedge clk) begin a = b; end

C. always @(clk) begin a <= b; end

D. always @(posedge clk) a = b;

Solution

1. Step 1: Identify the correct event control

   The always block should trigger on the positive edge of the clock: posedge clk.

2. Step 2: Use non-blocking assignment inside the block

   Non-blocking assignment uses <= to update signals sequentially.

3. Final Answer:

   always @(posedge clk) begin a <= b; end -> Option A

4. Quick Check:

   Non-blocking inside posedge always block = correct syntax [OK]

Hint: Use '<= ' inside 'always @(posedge clk)' for sequential logic [OK]

Common Mistakes:

• Using blocking '=' instead of '<=' in sequential blocks
• Triggering always block on wrong event
• Missing begin-end for multiple statements

3. What is the output of this Verilog code snippet after one positive clock edge?

reg a, b;
always @(posedge clk) begin
  a <= b;
  b <= 1'b1;
end

Assuming initial values a=0 and b=0.

medium

A. a=0, b=1

B. a=1, b=1

C. a=0, b=0

D. a=1, b=0

Solution

1. Step 1: Understand non-blocking update behavior

   Non-blocking assignments update all variables simultaneously after the always block finishes.

2. Step 2: Trace values before and after clock edge

   Before clock edge: a=0, b=0.
   At clock edge: a gets old b (0), b gets 1.
   After update: a=0, b=1.

3. Final Answer:

   a=0, b=1 -> Option A

4. Quick Check:

   Non-blocking updates use old values simultaneously [OK]

Hint: Non-blocking uses old values simultaneously on clock edge [OK]

Common Mistakes:

• Thinking 'a' updates immediately with new 'b'
• Confusing blocking and non-blocking timing
• Ignoring initial values

4. Identify the error in this Verilog code snippet:

always @(posedge clk) begin
  a = b;
  b = 1'b1;
end

where a and b are registers updated on clock edge.

medium

A. Missing sensitivity list for combinational logic

B. Using non-blocking assignments outside always block

C. Incorrect data types for registers

D. Using blocking assignments instead of non-blocking in sequential logic

Solution

1. Step 1: Check assignment type in clocked always block

   Sequential logic triggered on clock edges should use non-blocking assignments (<=), not blocking (=).

2. Step 2: Understand consequences of blocking in sequential logic

   Blocking assignments cause race conditions and incorrect simulation of flip-flops.

3. Final Answer:

   Using blocking assignments instead of non-blocking in sequential logic -> Option D

4. Quick Check:

   Sequential logic needs non-blocking assignments [OK]

Hint: Use '<=' in clocked always blocks to avoid race conditions [OK]

Common Mistakes:

• Using '=' in clocked always blocks
• Ignoring race condition risks
• Confusing sensitivity lists

5. You want to design a 2-bit register that updates both bits simultaneously on the rising edge of clk. Which code snippet correctly uses non-blocking assignments to avoid race conditions?

A) always @(posedge clk) begin
  bit0 <= bit1;
  bit1 <= bit0;
end

B) always @(posedge clk) begin
  bit0 = bit1;
  bit1 = bit0;
end

C) always @(posedge clk) begin
  bit0 <= bit1;
  bit1 <= bit1;
end

D) always @(posedge clk) begin
  bit0 = bit0;
  bit1 = bit1;
end

hard

A. Assigns bit1 to both bits, losing original bit0

B. Uses blocking assignments causing race conditions

C. Swaps bits correctly using non-blocking assignments

D. No change in bits, blocking assignments used

Solution

1. Step 1: Analyze non-blocking assignments in Swaps bits correctly using non-blocking assignments

   Swaps bits correctly using non-blocking assignments uses non-blocking assignments to swap bit0 and bit1 simultaneously, avoiding race conditions.

2. Step 2: Compare with other options

   Uses blocking assignments causing race conditions uses blocking assignments causing race conditions.
   Assigns bit1 to both bits, losing original bit0 overwrites bit0 and bit1 incorrectly.
   No change in bits, blocking assignments used does not change bits and uses blocking assignments.

3. Final Answer:

   Swaps bits correctly using non-blocking assignments -> Option C

4. Quick Check:

   Non-blocking swaps bits simultaneously [OK]

Hint: Use non-blocking to swap registers simultaneously on clock edge [OK]

Common Mistakes:

• Using blocking assignments in sequential logic
• Overwriting values before reading them
• Not updating both bits simultaneously