cc @petrochenkov: the above failure is due to the unexpected_cfg lint fire-ring on #[cfg(target_pointer_width = "8")] as "8" is not an expected value. I could add "8" to the list of extra values for std but I wonder if it wouldn't be better to set the list of target_pointer_width to 8, 16, 32, 64 and 128 (the current one are: 16, 32, 64) in the compiler instead of relying on the built-in targets for it ?

target_pointer_width = "8" looks like a mistake, I'm not aware of any out-of-tree targets using it, and there's a bunch of places in the compiler that produce asserts on any pointer width except for 16 | 32 | 64.

Just in case you haven't seen that it was ready for another review.

@rustbot ready

@bors r+ rollup=iffy

📌 Commit 1739793 has been approved by Mark-Simulacrum

⌛ Testing commit 1739793 with merge 21b0325...

☀️ Test successful - checks-actions
Approved by: Mark-Simulacrum
Pushing 21b0325 to master...

Finished benchmarking commit (21b0325): comparison url.

Summary: This benchmark run did not return any relevant results.

If you disagree with this performance assessment, please file an issue in rust-lang/rustc-perf.

@rustbot label: -perf-regression