steakhal wrote: I was initially surprised that the clang CFG was not changed in the PR. This made me suspicious. Later realized that since this change was implemented in terms of the `BreakStmt`, everything should work out of the box. I started playing with it and everything looked correct to me. In the end I decided to add a test pinning the CFG shape for some unnaturally complicated sample code I used for testing. Hopefully that's not going to become a churn long term, but we will see.
I thought I'll see yet another nail in the Clang CFG in favor of the CIR based structural CFG (that I'm really looking forward), because we have a tendency in clang to treat the Clang CFG as a second citizen as it's not used for Codegen, so it has many flaws lurking. Luckily, this PR have not made that situation worse. Thank you. FYI I have not looked at the PR itself, only checked that this should work for Clang CFG, Dataflow, and the Clang Static Analyzer. https://github.com/llvm/llvm-project/pull/152870 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
