dwblaikie wrote: How'd this work before your recent changes, then - when each repeated query would get one level further down in the nesting? How'd that work given the clang limitations you're describing?
In any case, the extra clang requirements here seem like they might be of uncertain cost/maintainability (if it's only updating one place that everyone calls through - great, but if it's updating multiple callers, and the risk that new callers miss this handling - that seems like a maintainability problem) which is worrying. So, I'd be uncertain about that direction without more info - and with that info, if it is just one codepath, yeah, maybe it's quick enough to address the regression. But if it'd require substantial reengineering to clang to get this working - this seems an unfortunate regression to carry in the interim, and it might be worth reverting to the previous (differently buggy, but at least work-around-able) behavior. And if this would require updating many callers in clang (unless such updates include an API change that would make it difficult to accidentally introduce a new caller that didn't handle things correctly for lldb), I'd worry about the future stability of such a change & might be inclined towards the less efficient "search all the DWARF" thing. https://github.com/llvm/llvm-project/pull/77029 _______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
