Coverage Extensibility in SystemVerilog-2023
Quite a while back I wrote a series of posts about coverage extensibility in SystemVerilog. In the first post we looked at how to use policy classes to implement extensible coverage collectors, whereβ¦
After a long hiatus, I finally wrote a new blog post: blog.verificationgentleman.com/2026/03/01/c... . This one's about the "embedded covergroup inheritance" feature added to SystemVerilog by IEEE 1800-2023. Is what I wrote in 2015 about coverage extensibility now obsolete? Read the post to find out
01.03.2026 21:52
π 1
π 0
π¬ 0
π 0
I joke, but in all seriousness, I am an AI optimist. It's just going to take a bit more effort from us digital verification folks.
15.02.2026 15:47
π 0
π 0
π¬ 0
π 0
Who says @GitHubCopilot can't be used for underrepresented languages like e (IEEE 1647)?
I asked it to fix this error in #Specman and it managed to do it! It only took a few iterations of trying to edit the wrong file, removing <' markers from working code and using HTML brackets before succeeding.
15.02.2026 15:47
π 0
π 0
π¬ 1
π 0
@GitHubCopilot made my day again by flagging some code as matching similar code, which actually turned out to be from one of my repos. Well, one of my repos and that of someone who copied it.
Had a look at the link. Seems like the author is η₯η§δΊΊ, which Google translates as "mysterious man". π
10.02.2026 20:32
π 0
π 0
π¬ 0
π 0
For all the talk about covergroup inheritance in #systemverilog that one can find online since 2023, it seems that EDA vendors couldn't really be bothered to implement it.
08.02.2026 17:07
π 1
π 1
π¬ 0
π 0
I found a quote that very summarizes one of my main reasons for preferring Git over other version control systems:
"Git isn't just a version control system, but part of a broader, highly integrated toolset that many teams rely on daily. ".
27.08.2025 07:32
π 0
π 0
π¬ 1
π 0
That feeling when GitHub Copilot warns you that it found similar code to what it generated and then it points you to your own repo. π
20.07.2025 17:39
π 0
π 0
π¬ 0
π 0
E.g. VS Code has doesn't like symlinks for indexing or refreshing Git status.
22.03.2025 07:29
π 0
π 0
π¬ 0
π 0
Absolutely necessary for IT to change $HOME from `/home/<user>` to `/<company-ticker>home/<user>` (where `/<company-ticker>home` is a symlink to `/home`). Absolutely necessary for all the little issues it causes with various programs (like VS Code) that don't like to traverse symlinks
22.03.2025 07:29
π 0
π 0
π¬ 1
π 0
Thanks, I think the same as you. What was confusing in the initial post was that all the other elements in the list are positive and that you also used ">shared ownership". It made it sound like code ownership is better.
18.02.2025 21:46
π 0
π 0
π¬ 0
π 0
Could you elaborate a bit on "code ownership" vs. "shared ownership"?
17.02.2025 07:00
π 0
π 0
π¬ 1
π 0
Also stealing "edit, pray, debug", fits so well with my industry.
08.02.2025 10:52
π 1
π 0
π¬ 0
π 0
Just finished learning how to build my first Astro blog! Check it out at docs.astro.build
via @astro.build
02.02.2025 17:44
π 1
π 0
π¬ 0
π 0
I was doing some productive procrastination by going through the tutorial for @astro.build and I have to say that it's absolutely fantastic. It introduces concepts progressively, let's the user make mistakes and then shows how to fix them, and is just all around easy to follow.
02.02.2025 15:46
π 1
π 0
π¬ 0
π 1
Yet again, I run into the problem that a crappy #semiEDA tool wrapper can't properly handle multi-token arguments (i.e. `wrapper --tool-arg 'some value'` leads to something like `tool --arg some value`). I'm getting pretty sick and tired of running into this again and again and again.
30.01.2025 12:21
π 0
π 0
π¬ 0
π 0
Why is careful commit naming useless on some teams? (I'm guessing this doesn't mean just using "upd" or "change" as a commit message.)
24.01.2025 07:11
π 0
π 0
π¬ 0
π 0
True dat! Same also applies to test benches. Still, somehow the industry has too much of a fondness for code generation, usually as a first resort.
17.01.2025 07:53
π 0
π 0
π¬ 1
π 0
Sometimes while having to work on projects developed by various "experts", I look at the screen and think "with code like this, who needs enemies?".
14.01.2025 17:41
π 0
π 0
π¬ 0
π 0
I just used `git bisect` for the first time to find a broken commit. I feel like a wizard!
03.01.2025 16:27
π 1
π 0
π¬ 0
π 0
Makes me wonder why there isn't a "SystemVerilog" group, like there is for SystemC.
27.12.2024 17:45
π 1
π 0
π¬ 0
π 0
Where does one normally ask questions to clarify contents of the SystemVerilog standard? StackOverflow doesn't feel like the best place for this somehow. On the Accellera forums there's the "UVM SystemVerilog Discussions" section, but that should be for UVM related stuff.
27.12.2024 17:45
π 2
π 0
π¬ 1
π 0
It's time for me to also jump on the Bluesky bandwagon.
24.12.2024 13:03
π 1
π 0
π¬ 0
π 0