AaronBallman wrote: I have no objections in theory, but in practice I remain concerned about the flakiness of lldb's tests for the impacts on Clang reviewers.
There are timeout failures, but they might be limited to Windows and thus not a problem with your PR (https://lab.llvm.org/buildbot/#/builders/219/builds/11819 and https://lab.llvm.org/buildbot/#/builders/219/builds/11810 as examples). But then there are failures like: https://lab.llvm.org/buildbot/#/builders/17/builds/53939 -- this same stack trace seems to pop up in other failures and looks to be unrelated to changes in the change set, such as https://lab.llvm.org/buildbot/#/builders/68/builds/75367 or https://lab.llvm.org/buildbot/#/builders/68/builds/75698 -- different assertion, but like above, seems to be spurious and unrelated to changes in the changeset, such as https://lab.llvm.org/buildbot/#/builders/68/builds/75666 Perhaps one way forward is for lldb tests to run on lldb changes for a few months before enabling them on Clang changes? Then the lldb folks can either report back that things are in pretty good shape from their experience, or we'll maybe have a better idea of how to proceed. That said, if others aren't as concerned about test flakiness, I won't block progress on this PR. https://github.com/llvm/llvm-project/pull/94208 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
