labath wrote: > Makes sense, LGTM. > > If the amount of work here is so small, and the granularity was too fine for > creating separate tasks, I'd argue that it's too fine for progress reporting > as well. I think that's what you're saying too. There's probably multiple > ways of dealing with that, so I'm looking forward to the PR that addresses > that.
It is, and there are. I have a prototype which deals with that by updating progress after every X bytes of DWARF have been indexed. The thing I like about that is that it should give a relatively constant rate of updates regardless of whether you're using type units (many small units) or not (fewer large units). The part I don't like is that does gives a wildly different rate for split vs. non-split DWARF (as in the former case, the main DWARF unit is just a handful of bytes). I'm still thinking about alternatives... https://github.com/llvm/llvm-project/pull/118657 _______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
