bogner wrote: > oh i thought this change did preserve that since this test should fail if > such an error is produced? Or that was my thought.
IIRC the error was only happening when the clang -cc1 command with both -S and -emit-llvm present was actually run, so the -### command wouldn't produce the error on its own. That said, testing that something doesn't happen for a bug that was fixed long ago is kind of tricky business - this is probably good enough as is. https://github.com/llvm/llvm-project/pull/165743 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
