labath added a comment. Thank you for the support Jonas. I've quickly prototyped a patch (https://reviews.llvm.org/D47265) of how would this integrate with lit. The nice part about it is that it is very small. If you consider the inline test refactor it even has negative LOC. It would definitely have negative LOC impact if this would allow us to remote test driver parts from dotest.
The not-so-nice part about it is that is has more impact on test running time than I expected. For me the time to run the tests goes from 1m38s to 2m30s. I am not extremely worried about that as that is still pretty fast for me (*), but I don't know how that scales, so it could be that this makes a significant difference for people with slower machines. I encourage you to give it a try (you need to apply it together with this patch), and let me know what you think. (*) I should also point out that the competing solution of putting every test into it's own file would likely have similar impact, as this is mostly due to fixed startup cost of spinning up a python +lldb process for the test. I also hope that some of this slowdown can be reclaimed as we garbage collect parts of dotest that would become unneeded after this. https://reviews.llvm.org/D46005 _______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
