Fabian Ruffy

> Or we can just get rid `boost::format` so the problem will go away Would be nice, but quite tough because of the difference in formatting... not even sure how...

> > What about [fmt](https://github.com/fmtlib/fmt)? Will also be much easier to replace with `std::format` in the future. > > it is nice, yes. But probably we'd need to check /...

Bounded loops are a top-level agenda item of the P4 language working group for this year: https://docs.google.com/document/d/1XSgdXeG1UuF1FM_XAqxDrHeN4dHZWBnJPKVS6SnGNwM/edit#heading=h.q7bzcwglv61g Upstreaming this is a great idea and very useful. How the syntax will...

> Why is the `p4test` assumed to be always present? Or is it just that `p4test` does not have any dependencies that are also not dependencies of the "core" used...

The relevant issue for this problem is https://github.com/p4lang/p4c/issues/4420. This requires adding checks for every back end and code that depends on the control plane.

May be a duplicate of #4500.

Transform is a very expensive visitor. It can copy the entire IR AST if small leaf elements like Path Expressions or members are cloned.

Another thing that could lead to the memory blowup is the use of `postorder`. I have noticed that for complex programs using `postorder` can lead to a strange memory leak....

Should we create an umbrella issue for all these regressions? It looks like reverting the PR is the way to go. I can also confirm that I have not enabled...

We can do that, but the PR should add all the issue programs as samples, too. It should also keep the program bug that was fixed (maybe add it to...