[Enselic]

The test passes with a nightly toolchain from before the corresponding fix was merged (see #106157 (comment) for instructions), so it is unlikely that the test will catch a regression. So we might as well remove it, it seems.

Unblocks fixing #112504 by fixing the CI failure in #165210 (comment).

CC @LeSeulArtichaut who authored the PR that introduced the test (which I pinged 5 days ago, see #106157 (comment)). (AFAIK the reviewer compiler-errors does not want to be pinged any longer, so I'm not pinging.)

[rustbot]

r? @adwinwhite

rustbot has assigned @adwinwhite.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

Why was this reviewer chosen?

The reviewer was selected based on:

• Owners of files modified in this PR: compiler
• compiler expanded to 79 candidates
• Random selection from 23 candidates

[adwinwhite]

The test is missing a flag. You need to add -Zthir-unsafeck to trigger the ICE with the older nightly.

Edited: the flag is on by default now so the test doesn't need additional annotation.

[adwinwhite]

So I guess it's useful?

@rustbot author

[Enselic]

@adwinwhite Thanks for finding that 👍

I see now that it is mentioned in the issue that the original PR fixed. It looks like the test actually did not work (= catch regressions) the day it was merged (Jan 4, 2023), but it started working one day later, when -Zthir-unsafeck was enabled by default (Jan 5, 2024). I suspect this was known at the time, and it's not a problem in practice of course.

Anyway, now that I know how to reproduce, I'll make a proper fix in another PR, so I'll close this one. Thanks again.