charles-zablit wrote: > > It's not a good solution, I am working on a better one. This would however > > unlock the bots while still keeping STDIN and STDOUT support in lldb. > > These tests are skipped currently, right? So the bots aren't failing, they're > only skipping. I think in this case, it would be fine if the proper fix for > this takes some time.
`lldb-shell :: Driver/TestEmptyArgument.test` is failing on https://lab.llvm.org/buildbot/#/builders/141/builds/15535. The x86_64 bot does not seem to be impacted. I reproduce the failure at desk (x64). The lldb-dap tests that were enabled recently are passing however. > > I think the correct solution would be to close the ConPTY when we receive > > the EOF in `ConnectionGenericFile::Read`. > > So it's not closed in the `ProcessWindows` destructor but in > `ConnectionGenericFile`? Since the PTY is behind a `shared_ptr` already, > could both `ProcessWindows` _and_ `ConnectionGenericFile` keep a reference? > Then the PTY would be closed when both are done. I will take a look at this tomorrow with a rested eye, but I don't think this can work: the read thread is never interrupted unless we destroy the ConPTY. On the other hand, we can't kill the read thread unless we know all the data has been consumed. Ideally we would want to destroy the ConPTY when we receive "EOF" but that never seems to get sent. https://github.com/llvm/llvm-project/pull/182109 _______________________________________________ lldb-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
