r? @alexcrichton

(rust_highfive has picked a reviewer for you, use r? to override)

O.o This is a HUGE refactoring! 👏 👍 🍰

I also wonder how many PRs will be marked as unmergeable because of this... 😆

@barosl it's a... prerequisite for a prerequisite (TyCx<'a, 'tcx>) for split local/global type contexts.
I'm a bit worried that it doesn't help as much as I had hoped or that it will be controversial.

It's also going to rot if it doesn't get reviewed this week, but most changes were mechanical, so unless there are significant changes to middle::ty, it should be possible to rebase it manually.

r? @nrc (or other @rust-lang/compiler folks)

r? @nikomatsakis

☔ The latest upstream changes (presumably #26415) made this pull request unmergeable. Please resolve the merge conflicts.

I might remove the last two commits - if we can get something like '_ working (to replace &ty::ctxt<'tcx> with TyCx<'_, 'tcx>), even if it's feature-gated, the existing TyCx commits would be obsoleted (and merging them would make it harder to switch to '_ later).
Also see https://internals.rust-lang.org/t/flattening-the-contexts-for-fun-and-profit/2293.

@nikomatsakis I've removed the TyCx commits from this PR - they are now at https://github.com/eddyb/rust/commits/tycx-more-tcx-tcx.
If #26598 is merged, introducing TyCx will be way less painful (though it does mean a couple of days were wasted) - also see rust-lang/rfcs#1177.

@bors r+

📌 Commit ad66c21 has been approved by nikomatsakis

@eddyb this is nice!

⌛ Testing commit ad66c21 with merge 650566e...

☀️ Test successful - auto-linux-32-nopt-t, auto-linux-32-opt, auto-linux-64-nopt-t, auto-linux-64-opt, auto-linux-64-x-android-t, auto-mac-32-opt, auto-mac-64-nopt-t, auto-mac-64-opt, auto-win-gnu-32-nopt-t, auto-win-gnu-32-opt, auto-win-gnu-64-nopt-t, auto-win-gnu-64-opt