Revert "trait_selection: Keep type-op region constraints in borrowck"

[Dnreikronos]

This reverts #161423.

Boxy pointed out the problem after it landed: the code I touched there runs a lot, and I made it do the new region constraint work every time, even when assumptions-on-binders is off and the constraint is just true. So in the common case you end up paying for building and carrying around something nobody ever looks at. Fair point, and honestly I should have seen it myself while writing it. I was thinking about getting the correctness right and not really about how often those functions get called.

The plan is to revert now and then re-land the same thing with the new logic behind checks for whether AoB is actually on, so we can get a perf run and look at real numbers instead of guessing.

There is nothing clever in this PR. It is just git revert -m 1 on the rollup merge, so the diff is the exact opposite of what went in. It applied with no conflicts on current main and x check compiler is fine, which I wanted to confirm because #162238 landed after this one and touches some of the same files.

r? @BoxyUwU

[rustbot]

Some changes occurred to the core trait solver

cc @rust-lang/initiative-trait-system-refactor

[rustbot]

BoxyUwU is currently at their maximum review capacity.
They may take a while to respond.

[BoxyUwU]

@bors r+ rollup=never

thanks for posting this, sorry for not catching that it might be perf sensitive

[rust-bors]

📌 Commit 393c127 has been approved by BoxyUwU

It is now in the queue for this repository.

[rust-bors]

💔 Test for c8976fe failed: CI. Failed job:

• auto - dist-aarch64-llvm-mingw (web logs, enhanced plaintext logs)

[JonathanBrouwer]

@bors retry

[rust-log-analyzer]

A job failed! Check out the build log: (web) (plain enhanced) (plain)

Click to see the possible cause of the failure (guessed by this bot)

[JonathanBrouwer]

@bors retry

[rust-bors]

❗ You can only retry pull requests that are approved and have a previously failed auto build.

Hint: If you wanted to retry pull request CI instead, push the latest commit again, or close and then reopen this PR.

[rust-bors]

☀️ Test successful - CI
Approved by: BoxyUwU
Duration: 3h 10m 37s
Pushing 315ecf4 to main...

[github-actions]

What is this?

This is an experimental post-merge analysis report that shows differences in test outcomes between the merged PR and its parent PR.

Comparing a36d05e (parent) -> 315ecf4 (this PR)

Test differences

Show 8 test diffs

Stage 0

• infer::solver_region_constraints::tests::true_constraint_keeps_query_response_empty: pass -> [missing] (J1)

Stage 1

• infer::solver_region_constraints::tests::true_constraint_keeps_query_response_empty: pass -> [missing] (J0)

Additionally, 6 doctest diffs were found. These are ignored, as they are noisy.

Job group index

• J0: test-aarch64-apple-2, test-aarch64-apple-macos-26-2, test-aarch64-gnu, test-aarch64-gnu-llvm-21-2, test-aarch64-msvc-2, test-i686-gnu-2, test-i686-gnu-nopt-2, test-x86_64-gnu, test-x86_64-gnu-llvm-21-1, test-x86_64-gnu-llvm-22-1, test-x86_64-gnu-nopt, test-x86_64-gnu-stable, test-x86_64-mingw-2, test-x86_64-msvc-2
• J1: test-x86_64-gnu-llvm-21-3, test-x86_64-gnu-llvm-22-3

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard 315ecf4a939def16631c2b25c3782ad67fc22160 --output-dir test-dashboard

And then open test-dashboard/index.html in your browser to see an overview of all executed tests.

Job duration changes

1. test-x86_64-gnu-miri: 47m 31s -> 1h 23m (+76.1%)
2. test-armhf-gnu: 1h -> 1h 38m (+62.4%)
3. test-x86_64-gnu-stdlib-semver-check: 11m 30s -> 17m 45s (+54.3%)
4. dist-ohos-aarch64: 1h 3m -> 1h 23m (+32.2%)
5. test-i686-gnu-2: 1h 44m -> 1h 10m (-31.9%)
6. test-x86_64-gnu-parallel-frontend: 1h 31m -> 1h 59m (+31.4%)
7. test-i686-gnu-1: 2h 31m -> 1h 48m (-28.5%)
8. test-x86_64-gnu-gcc-core-tests: 12m 39s -> 16m 8s (+27.5%)
9. dist-x86_64-mingw: 2h 8m -> 2h 42m (+26.3%)
10. dist-arm-linux-gnueabi: 1h 36m -> 1h 11m (-25.3%)

How to interpret the job duration changes?

Job durations can vary a lot, based on the actual runner instance
that executed the job, system noise, invalidated caches, etc. The table above is provided
mostly for t-infra members, for simpler debugging of potential CI slow-downs.

[rust-timer]

Finished benchmarking commit (315ecf4): comparison URL.

Overall result: ✅ improvements - no action needed

@rustbot label: -perf-regression

Instruction count

Our most reliable metric. Used to determine the overall result above. However, even this metric can be noisy.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-2.3%	[-6.2%, -0.2%]	167
Improvements ✅ (secondary)	-2.3%	[-6.3%, -0.1%]	181
All ❌✅ (primary)	-2.3%	[-6.2%, -0.2%]	167

Max RSS (memory usage)

Results (primary -0.8%, secondary -1.1%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	0.4%	[0.4%, 0.4%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.9%	[-1.5%, -0.5%]	11
Improvements ✅ (secondary)	-1.1%	[-2.1%, -0.9%]	5
All ❌✅ (primary)	-0.8%	[-1.5%, 0.4%]	12

Cycles

Results (primary -2.5%, secondary -3.5%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	1.5%	[0.5%, 3.4%]	3
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-2.7%	[-4.3%, -0.5%]	55
Improvements ✅ (secondary)	-3.5%	[-5.2%, -2.1%]	41
All ❌✅ (primary)	-2.5%	[-4.3%, 3.4%]	58

Binary size

This perf run didn't have relevant results for this metric.

Bootstrap: 481.833s -> 478.997s (-0.59%)
Artifact size: 403.27 MiB -> 403.72 MiB (0.11%)