jimingham wrote: That would require that the thread wouldn't have a direction itself, but the direction would be governed by whoever explained the stop and set up the next phase of execution. Since all the threads need to work in lock step, that probably means that `thread step-over --direction <value>` needs to not only push a "thread plan step-over" with the right direction on the current thread, but all the other threads get a "thread plan go in a direction" plan that coordinates with the plan that's moving the process along about when it should be done.
That's linking the thread plans from different threads, which we haven't much done yet, but I think that's better than trying to hold some state outside the thread plan stack and some in it. https://github.com/llvm/llvm-project/pull/112079 _______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
