`test tests/ui/entry.rs` failing on `nightly-2021-10-10`

[timrobertsdev]

On Linux, the ui test for the entry attribute are failing due to a warning being emitted before the expected output.

Output:
WARN rustc_metadata::locator no metadata found: failed to mmap rmeta metadata: '$WORKSPACE/target/tests/target/x86_64-unknown-linux-gnu/debug/deps/libuefi_macros_tests-ec1cf1694a730703.rmeta'

I have confirmed that $WORKSPACE/target/tests/target/x86_64-unknown-linux-gnu/debug/deps/libuefi_macros_tests-ec1cf1694a730703.rmeta exists, but it is an empty file. Not quite sure what's going on yet, as I'm not too familiar with the inner workings of cargo and rustc.

This is also failing on Windows using the MSVC toolchain, but with a different error.

Output:
ERROR: failed to execute cargo: %1 is not a valid Win32 application. (os error 193)

There doesn't appear to be anything in the last few commits that would cause this; I'm leaning towards it being a bug in rustc/cargo. It's getting late here now, I'll try some different nightly versions tomorrow to figure out what's going on.

Steps to reproduce:

1. clone master
2. cd uefi-rs/uefi-test-runner
3. ./build.py test

Environment (Linux):

• WSL2 (Ubuntu 20.04, 5.10.60.1-microsoft-standard-WSL2), Fedora 34 (5.14.9-200.fc34.x86_64)
• rustc --version: rustc 1.57.0-nightly (41dfaaa3c 2021-10-10)

Environment (Windows):

• Windows 11 (build 22000.194)
• rustc --version: rustc 1.57.0-nightly (41dfaaa3c 2021-10-10)

[nicholasbishop]

Looks like rust-lang/rust#89795

[nicholasbishop]

I've temporarily disabled the test with #300. Will keep this bug open as a reminder to re-enable the test once the underlying issue is fixed.