https://github.com/DavidSpickett commented:
For it's stated purpose, this looks fine. > Depends on https://github.com/llvm/llvm-project/pull/207764. For both these PRs, I need to see some functional changes before we proceed with changes to the test suite. I am mainly wondering whether we're actually going to be integrating Flang or just using it as a compiler. This PR doesn't seem to mind where Flang comes from, but the other one requires us to build it. Roughly there's 3 levels to this, where each one will get less coverage because the requirements increase each time: * Stuff you can test with synthetic inputs * Stuff you just build with an existing compiler and it's so basic the compiler just has to be from the current decade * You need to link against flang And if the work has not reached the third one, I'd rather not require building flang. A buildbot might build flang because it wants to but I don't want to make anyone else pay that cost. If the very first functional change is going to link against flang then ok, we pay the cost, so let's see those changes before we do anything with the test suite. You can do a stacked PR for that or just comment to say that it's based on these 2 existing PRs and include all the changes in that one, so that tests actually run. https://github.com/llvm/llvm-project/pull/208298 _______________________________________________ lldb-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
