Pitfalls for circuit girls

The Circuit Girl

Some time ago I heard about a video with FPGA design book recommendations by Jeri Ellsworth, who is also known as the Circuit Girl. I admit that the concept of circuit girls triggered my imagination. Somehow I feel like there should be more of those. Consequently, I checked out her video without further delay.

Here is a successful, charming engineer presenting her favourite FPGA design books. She definitely had my ear. In the video, she refers to “programming” an FPGA in VHDL or Verilog. I like that. Her top pick is a book called HDL Chip Design by Douglas Smith, informally know as the blue book. She describes it as a must-have. Needless to say, I ordered it immediately. A used copy though, because it is an expensive beast.

Checking out the blue book

The blue book is rather intimidating, if only for its size. One definite plus is that it discusses both VHDL and Verilog. Moreover, the foreword promises that it will be “a beacon in your quest for perfect HDL design”. While glancing over it, I got the impression that it contains a lot of useful information. However, my first action was to look for errors and bad advice. That is what I always do when I get my hands on a book about HDL design. I suppose the reason is a mixture of bad character and experience. Anyway, I did not have to look very hard.

On page 174, the author demonstrates how to describe flip-flops. According to the book, a simple D flip-flop without reset would be described in Verilog like so:

module ff(clock, a, y);
    input clock;
    input a;
    output y;
    reg y;
    always @(posedge clock)
        y = a;

In a stand-alone simulation, your simulator may tell you that this is indeed a flip-flop, as may your synthesis tool. However, let me be clear: it is not. At least not always. Only when you are lucky.

The problem

A flip-flop is a building block for more complex circuitry. For example, consider a simple two-stage delay line:

module delay_line(clock, a, y);
    input clock;
    input a;
    output y;
    wire y1;
    ff dff1(clock, a, y1);
    ff dff2(clock, y1, y);

With the models as presented, this model may behave as the intended two-stage delay line, or as an unintended single stage. There is nothing in Verilog to prefer one behavior over the other. It depends on your simulator brand. Or version. Or on source code order. Or on some random, unpredictable event. Both outcomes are equally valid. This is nondeterminism in practice.

To see the problem more clearly, consider the following code, which is equivalent from the point of view of the Verilog scheduling algorithm:

module delay_line(clock, a, y);
    input clock;
    input a;
    output y;
    reg y;
    reg y1;
    always @(posedge clock)
        y1 = a;
    always @(posedge clock)
        y = y1;

If you simulate this on any Verilog simulator, and experiment with the order of the always blocks, you will most likely see the issue.

One problem with nondeterminism, as with bugs in general, is that it is impossible to prove its absence by simulation. To complicate matters further, your simulator may hide the nondeterminism in the original example. The reason is that some simulators, including the original Verilog simulator and its descendants from Cadence, handle ports in a special way. However, don’t be misled by this. The Verilog standard scheduling algorithm has no special provisions for ports . Other simulators, such as Synopsys VCS, have exploited this characteristic by implementing “optimizations” that will reveal the nondeterminism.

As you may recall from previous posts, the problem with the flip-flop code is the use of a blocking assignment “=” for synchronous communication. This coding error is made systematically throughout the whole blue book. The fix would be to use non-blocking assignments “<=” in such cases instead.

According to Jeri Ellsworth, the blue book is very popular. This suggests that many companies follow its coding style. As a result, there may be a lot of fragile Verilog legacy code out there: code that has “always worked” but that suddenly will stop working with a new version or brand of the Verilog simulator.

I am puzzled by the fact that the blue book was never corrected. The author certainly was aware of non-blocking assignments. In some examples he uses them for their parallel semantics in a single thread of code. Perhaps he was not aware of their main purpose, but others could have told him. According to the foreword, “[the book’s] accuracy has been verified by leading industry experts”. Perhaps the issue was not yet clear to them in the first edition of 1996. However, my copy of the book is the ninth printing from July 2001. It seems there was ample expertise and time available for corrections, but for some reason they were not made. I have no indication of later editions that may have fixed the problem.

How to fix it

One factor may be that the required corrections are not trivial. We certainly don’t want to substitute all blocking assignments by non-blocking assignments, as their semantics are completely different. Only assignments used for communication should be considered. However, we don’t want to subsitute all of those either. For combinatorial logic, there is no issue with nondeterminism, and Verilog designers typically prefer blocking assignments in this case. In summary, we only want to substitute assignments used for communication and controlled by a clock edge.

What I have just described is a typical example of a refactoring. By definition, a refactoring is a code transformation intended to improve the code while keeping its intended behavior intact. As can be seen from this example, interesting refactorings are often non-trivial. No matter how desirable, manual refactorings are often avoided because there is a significant risk to introduce new errors in the process.

To make interesting refactorings practical, they must be automated, by using a development environment that understands the programming language. In a Verilog IDE product, a Fix Nondeterminism refactoring would be at the top of my list of desired features. It could be used to fix the blue book automatically. In the development environment, it would be available as a code warning with an associated quick-fix. This would avoid the issue at the root.


As you have guessed already, all these considerations are irrelevant for VHDL. Any VHDL example from the blue book will behave identically on any compliant VHDL simulator. By comparison, the cost of Verilog’s nondeterminism seems disproportionally high. Why did the Verilog language designers not avoid it like VHDL does? Did they underestimate the issues? Or did they perhaps have a good reason that we have not appreciated yet? We will address these questions on the next occasion.


When reading this article,

When reading this article, I'm happy that I did choose VHDL instead of Verilog (by accident only!).

I'm curious about the

I'm curious about the reasons. Of course Verilog was truly revolutionary when it came out, but I wonder why it did not evolve and correct itself like so many other technologies?

I suppose they did not want to break backwards compatibility? Is Verilog the Win32 API of the hardware world?

How to fix it.

I would put both variable assignments into a single always block.
Now the statement order will determine how the bits bounce.
I reserve <= non-blocking assignments for I/O.
Notice that modelsim will also combine blocks/processes with the same clk, reset signature.

-- Mike Treseler

Make that "How to hack it."

I don't doubt that Jan is correct about how to fix verilog.
My point is that a single block module is a viable atom until the fix is in. -- Mike Treseler

The fix is not about RTL


I assume you refer to this post.

The style you advocate is perfect for RTL and I use a similar (slightly less pure) one. The Verilog community as a whole doesn't like it, but that is a different subject, for many future posts :-)

However, this style doesn't avoid nondeterminism in general, for example for behavioral event-driven models without a clock. That is the target of the proposed fix.

The fix is not about RTL

OK. I see.
It is interesting that even with a deterministic VHDL testbench there can still be user confusion on interface issues like handling of zero setup setup time. --MT

Verilog's ulterior motives

I once had a talk with one of my professors, who was part of the committee that initally specified the Verilog language. I had been a VHDL designer for years, and those lectures were full of small things that were bothering me about Verilog.

I made notes of several of those details, and at the end of one class I grabbed him and started asking the first question in my list, hoping to hear (like you) that the Verilog designers had some misterious ulterior motives.

His answer shocked me so much that I left most of my questions unanswered: "Because we didn't know better".