-
Notifications
You must be signed in to change notification settings - Fork 12.8k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
lldb debuginfo test failure, hash_set/hash_map of size 0 instead of expected size 4? #83891
Comments
Incidentally, when I tried to run the test by hand (which admittedly I might have messed up in some way), I saw this: Click to see some details about my attempt to run `rust-lldb` on a binary.
It seems pretty suspicious that lldb was reporting (But that also doesn't actually seem to be the behavior that x.py is observing, so that might be user error, or a separate bug entirely.) |
I'm not sure if you've seen #81813, but the lldb tests only run on macOS on CI. It looks like you are maybe using linux, so this is exercising untested code. In particular, pretty-std-collections is disabled on macOS, so it is not tested anywhere. It's not too surprising that it may have bitrotted. I'm also curious, are you using lldb installed on your system? Is it the same version of LLVM that rustc was built with? |
Yeah I knew the debugger test suite i only enabled in certain contexts, but I couldn't remember the details. My runs of I should probably see if things change if I set (In any case, I'd say it remains a suboptimal developer experience to have x.py test fail in this manner.) |
Agreed, I ran into a similar problem (#81814). As long as the test is disabled in CI, it'll probably keep breaking unfortunately. I'm pretty certain I tested #81834 locally, and it worked, so somehow it broke again. Maybe @ortem has some insight? |
I can bisect, thanks for the link @ehuss ! |
Thanks @ortem, that was quick! And please don't feel obligated to investigate when I ping. It was just a wild guess if you might know what was up. |
@ehuss No problem, please feel free to mention me in case of any debugger pretty-printers problems! I'm working on the debugging support in IntelliJ Rust and I'm interested in fixing issues like this ASAP. |
I investigated the problem with a clear mind and found out that LLDB pretty-printer was updated improperly in #77566 (although GDB pretty-printer is OK). Please check the new PR. @ehuss It would be so great to have LLDB pretty-printers tests on CI. In IntelliJ Rust, we run them on Linux. We use |
I tried this code:
I expected to see this happen: successful test suite run
Instead, this happened: debuginfo test failure for pretty-std-collection.rs, saying that an expecting output from the debugger was not found.
The specific line it seemed be this one:
Click to see transcript that includes the full stdout/stderr output as reported by `x.y` + `compiletest`
The text was updated successfully, but these errors were encountered: