felipepiovezan wrote: > Maybe a Doxygen comment explaining why you get the API lock back but not the > stop lock would address your concerns?
This is exactly how this started, actually. I felt like I should document why we get one lock back but not the other, and explain why such a method is useful. But then I found my self writing too much and realized that, rather than explaining the problem, I could define the problem away entirely by having a resume method. > It also seems like it should be possible to query the target (but probably > not the process) while holding the API lock, but after giving up the stop > lock. IMO this type of hypothetical is how we ended up with the design that allows for a bunch of behaviors to happen, and yet none of the users of the design needed those behaviors. Which is another way of saying that if, in the future, we needed what you described, we could just implement it. That said, I can see why a resume a method in the execution context could feel off, so let's just leave things as is. https://github.com/llvm/llvm-project/pull/152020 _______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
