Fwiw I’ve seen cases where tests have passed even though they shouldn’t have — the functionality being tested was broken. The one that comes to mind was where we were doing a backtrace and then checking that it matched the regex “main\(argc=3” to make sure the local variable argc had the correct value. But the actual backtrace was more like mian(argc=3751589203, ...). I.e. a garbage value.
This test passed for months this way until an unrelated change caused argc to change to a different junk value in the backtrace This isn’t necessarily the case here, but something to keep in mind. On Mon, Jul 16, 2018 at 10:10 AM Stella Stamenova via Phabricator < revi...@reviews.llvm.org> wrote: > stella.stamenova added a comment. > > I'll spend some time looking into this today, but with commit > 0fa537f42f1af238c74bf41998dc1af31195839a variables.test passes. Then with > commit d9899ad86e0a9b05781015cacced1438fcf70343, the test fails. There are > clearly a couple of other commits in that range, but they are a lot less > likely to have caused the failure. > > > Repository: > rL LLVM > > https://reviews.llvm.org/D49018 > > > >
_______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
