tests/ui/associated-consts/issue-93775.rs seems to be flaky

[jieyouxu]

I was modifying the compiler for unrelated changes on x86_64-pc-windows-msvc, and the tests/ui/associated-consts/issue-93775.rs ui test seemingly randomly failed, a rerun with nothing changed made the test pass, and now I can't reproduce it easily.

[jieyouxu]

cc #93775

[jieyouxu]

I don't have a good solution to this, having the test being flaky is not ideal, but neither is outright deleting this because this is a regression test for the original issue that segfaulted.

[DianQK]

Is it reasonable to add //@ retry=1 directive?

[jieyouxu]

Is it reasonable to add //@ retry=1 directive?

I haven't figured out why this particular test fails on msvc sometimes, and I think since it's currently disabled on msvc, it's better to figure out why it was failing in the first place.

[DianQK]

Is it reasonable to add //@ retry=1 directive?

I haven't figured out why this particular test fails on msvc sometimes, and I think since it's currently disabled on msvc, it's better to figure out why it was failing in the first place.

Here is a recursive function, and I think this issue remains unresolved.

rust/compiler/rustc_middle/src/ty/print/mod.rs

Lines 166 to 169 in 73c278f

	return self.path_generic_args(
	|cx| cx.print_def_path(def_id, parent_args),
	args,
	);