labath added a comment. The fix seems simple enough, but Jim needs to say whether this is the right way to fix this bug, as I am not sure about what are our assumptions about Breakpoint object lifetimes.
================ Comment at: packages/Python/lldbsuite/test/functionalities/breakpoint/breakpoint_command/main.c:13 { + printf("Observable side effect\n"); return 0; // Set break point at this line. ---------------- Why did you need to add this? This seems like something that could easily be removed/reshuffled in the future (breaking your test, if it depends on it). ================ Comment at: source/Breakpoint/BreakpointList.cpp:120-121 + auto bp = *pos; + if (bp->AllowDelete()) { + bp->ClearAllBreakpointSites(); + pos = m_breakpoints.erase(pos); ---------------- Don't we need to do the same thing in the other "remove" functions (`Remove`, `RemoveAll`). https://reviews.llvm.org/D45554 _______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits