Codegen / Intel® oneAPI & Fornux C++ Superset (New Deterministic Memory Manager)


As some of you might know already, “Fornux C++ Superset” is a source-to-source compiler, but it can easily be integrated into Codegen / Intel® oneAPI to provide the best solution customers can have with all other components you already provide in order to fix memory management issues.

Here are comparison matrices since “Fornux C++ Superset” proactively and implicitly fixes memory management issues once at compile-time, so there is no need for any extra reactive and explicit debugger which in turn cannot work with code that is cross-compiled on small embedded devices frequently used by military systems:

As you can imagine, the embedded software market where the UI apps don’t work also, is gaining important traction:

That being said, we hope to provide, through Codegen / Intel® oneAPI, the best solution in the industry, by integrating Fornux C++ Superset into Codegen / Intel® oneAPI.

Please find here our pitch deck for C++ Superset and AI Powershift:

Trials can be downloaded from here as well for all platforms:

And its documentation from here:



Is this using LLVM or Clang? Do you have some more information about the implementation?

I can’t really tell what is the point of your email just now but I’m curious to find out?


It rewrites the input C/C++ code and integrates the following headers:

Using clang-check.

It’s really top notch as I’ve been working on it for the last 10+ years, and there are single & multi threaded examples in the packages you can download.

So you understand the idea?

I can add also that Linus Torvalds always said that C++ “fixes the wrong problems”… Now he can’t complain anymore!

Sorry I didn’t fully answer your questions this week because I was in a rush but my point is to integrate “Fornux C++ Superset” into Intel(R) oneAPI to make the ultimate C++ compilation tool.

As you know, the core C/C++ memory management design is flawed and this is exactly what we are fixing by rewriting the source code using a static code analysis tool and integrating a predictable memory manager.

  1. As mentioned in the presentation, Google Chrome / Chromium / Brave is infested with cybersecurity issues, mostly based on bad memory management that can easily be fixed with “Fornux C++ Superset”. Facebook is also very sluggish after a while on the desktop as well as IP TV that suffers from memory leaks left and right.

  2. The Linux kernel and modules can ultimately be flawless upon only 1 compilation effort;

  3. Military systems use static memory to bypass this problem but this just creates different problems and the software can still crash leading to creating snowball effects;

  4. It is also used by our second Python to C++ source-to-source compiler called “AI Powershift”, to prevent Python apps from reverse engineering and to speed them up while at it.

So hopefully it is more clear now and you understand its design and our intends.