This test doesn't really make any sense anymore, it became broken a long time ago.

Simplify the path operation with `join`, clarify some of the names.

…xpr-attrs, r=davidtwco"

This reverts commit 57dad1d, reversing
changes made to 36316df.

…iper

Size optimize int formatting

Let's use the new feature flag!

This uses a simpler algorithm to format integers.
It is slower, but also smaller.
It also saves having to import the 200 byte rodata lookup table.

In a test of mine this saves ~300 bytes total of a cortex-m binary that does integer formatting.
For a 16KB device, that's almost 2%.

Note though that for opt-level 3 the text size actually grows by 116 bytes.
Still a win in total. I'm not sure why the generated code is bigger than the more fancy algo. Maybe the smaller algo lends itself more to inlining and duplicating?

Uplift `Relate`/`TypeRelation` into `rustc_next_trait_solver`

For use in the new solver. This doesn't yet uplift `ObligationEmittingRelation`.

r? lcnr

… r=petrochenkov

Don't warn on fields in the `unreachable_pub` lint

This PR restrict the `unreachable_pub` lint by not linting on `pub` fields of `pub(restricted)` structs and unions. This is done because that can quickly clutter the code for an uncertain value, in particular since the "real" visibility is defined by the parent (the struct it-self).

This is meant to address one of the last concern of the `unreachable_pub` lint.

r? ``@petrochenkov``

…t, r=jieyouxu

Remove `same-lib-two-locations-no-panic` run-make test

This test doesn't really make any sense anymore, it became broken a long time ago.

r? ``@jieyouxu``

…=jieyouxu

Crate loader cleanups

Minor cleanups I found while trying to understand how all of this works

Revert "Disallow ambiguous attributes on expressions" on nightly

As discussed in [today's t-compiler meeting](https://rust-lang.zulipchat.com/#narrow/stream/238009-t-compiler.2Fmeetings/topic/.5Bweekly.5D.202024-06-06/near/443079505), this reverts PR rust-lang#124099 to fix P-critical beta regressions rust-lang#125199.

r? ``@wesleywiser``

Opening as draft so that ``@wesleywiser`` and ``@apiraino,`` you can tell me whether you wanted:
1. a `beta-accepted` revert of rust-lang#124099 on nightly (this PR)? That will need to be backported to beta (even though rust-lang#126093 may be the last of those)
2. a revert of rust-lang#124099 on beta?
3. all of the above?

I also opened rust-lang#126102, another draft PR to revert rust-lang#124099 on beta, should you choose options 2 or 3.

…mpl, r=fmease

Improve Docs for `hir::Impl` and `hir::ImplItem`

Based on https://rust-lang.zulipchat.com/#narrow/stream/182449-t-compiler.2Fhelp/topic/.E2.9C.94.20Difference.20between.20.60hir.3A.3AImplItem.60.20and.20.60hir.3A.3AImpl.60.3F/near/442650915

r​? fmease