Compute lint_levels by definition

[fee1-dead]

Closes #95094.

cc @cjgillot

[rust-highfive]

r? @nagisa

(rust-highfive has picked a reviewer for you, use r? to override)

[fee1-dead]

cc @xFrednet this touches impl for lint expectations

[nagisa]

r? rust-lang/compiler

Not able to take on large reviews for the time being.

[cjgillot]

@bors try @rust-timer queue

[rust-timer]

Awaiting bors try build completion.

@rustbot label: +S-waiting-on-perf

[bors]

⌛ Trying commit 224cfded3fde57789375420501eb2146971eb807 with merge f2b76fa8b8a81c9059018568c196cf62f6ba4fdf...

[bors]

☀️ Try build successful - checks-actions
Build commit: f2b76fa8b8a81c9059018568c196cf62f6ba4fdf (f2b76fa8b8a81c9059018568c196cf62f6ba4fdf)

[rust-timer]

Queued f2b76fa8b8a81c9059018568c196cf62f6ba4fdf with parent c32dcbb, future comparison URL.

[rust-timer]

Finished benchmarking commit (f2b76fa8b8a81c9059018568c196cf62f6ba4fdf): comparison url.

Instruction count

• Primary benchmarks: mixed results
• Secondary benchmarks: mixed results

	mean1	max	count2
Regressions 😿 (primary)	14.5%	62.7%	94
Regressions 😿 (secondary)	23.8%	254.5%	75
Improvements 🎉 (primary)	-0.6%	-1.2%	12
Improvements 🎉 (secondary)	-0.7%	-1.3%	37
All 😿🎉 (primary)	12.8%	62.7%	106

Max RSS (memory usage)

Results

• Primary benchmarks: 😿 relevant regressions found
• Secondary benchmarks: 😿 relevant regressions found

	mean1	max	count2
Regressions 😿 (primary)	10.6%	26.3%	77
Regressions 😿 (secondary)	6.9%	16.6%	31
Improvements 🎉 (primary)	N/A	N/A	0
Improvements 🎉 (secondary)	N/A	N/A	0
All 😿🎉 (primary)	10.6%	26.3%	77

Cycles

Results

• Primary benchmarks: 😿 relevant regressions found
• Secondary benchmarks: 😿 relevant regressions found

	mean1	max	count2
Regressions 😿 (primary)	25.1%	86.5%	78
Regressions 😿 (secondary)	37.8%	259.4%	51
Improvements 🎉 (primary)	-2.5%	-2.8%	2
Improvements 🎉 (secondary)	N/A	N/A	0
All 😿🎉 (primary)	24.4%	86.5%	80

If you disagree with this performance assessment, please file an issue in rust-lang/rustc-perf.

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please indicate this with @rustbot label: +perf-regression-triaged along with sufficient written justification. If you cannot justify the regressions please fix the regressions and do another perf run. If the next run shows neutral or positive results, the label will be automatically removed.

@bors rollup=never
@rustbot label: +S-waiting-on-review -S-waiting-on-perf +perf-regression

Footnotes

1. the arithmetic mean of the percent change ↩ ↩² ↩³

2. number of relevant changes ↩ ↩² ↩³

[davidtwco]

Not exceptionally familiar with the lint level computation but this seems reasonable (when not considering the big perf regression), will re-assign to cjgillot though.

[davidtwco]

r? @cjgillot

[fee1-dead]

Yeah, I'm going to need some help regarding the perf hits here. @cjgillot could you take a look? Thanks.

[cjgillot]

You caught a bad case of incremental compilation invalidation.

When the incr. comp. engine tries to mark a query as green, it first checks whether all its dependencies are green. lint_levels_on depends on resolutions, which is always red.
As it can trivially mark lint_levels_on as green, the incr. comp. engine tries to recompute the query. The query takes a HirId but the incr. comp. engine does not know how to make a HirId from its Fingerprint. So it cannot recompute the query, and marks lint_levels_on as red.
As a consequence, all queries that call it (= pretty much everything) get recomputed.

There are 2 possibilities here:

1. compute lint_levels_on by definition: for a whole item/trait-item/impl-item/foreign-item at once, and make lint_levels_on take a LocalDefId;
2. teach the query system to force a query that takes a HirId.

(1) may need to a larger refactor of the levels computation. It will avoid the query call overhead (there are many more HirIds that LocalDefIds), but that overhead only appears as a ~1% regression, so that's probably fine.

(2) is a bit more experimental. For this, you need to:

• add a variant to rustc_query_system::dep_graph::FingerprintStyle;
• return it in the <HirId as DepNodeParams>::fingerprint_style in rustc_middle::dep_graph::dep_node;
• modify to_fingerprint and recover in that method have a bidirectional conversion Fingerprint <-> HirId (look at what LocalDefId does for instance).

[fee1-dead]

rebase; no work yet

[bors]

☔ The latest upstream changes (presumably #100426) made this pull request unmergeable. Please resolve the merge conflicts.

[rust-log-analyzer]

The job mingw-check failed! Check out the build log: (web) (plain)

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

configure: rust.debug-assertions := True
configure: rust.overflow-checks := True
configure: llvm.assertions      := True
configure: dist.missing-tools   := True
configure: build.configure-args := ['--enable-sccache', '--disable-manage-submodu ...
configure: writing `config.toml` in current directory
configure: 
configure: run `python /checkout/x.py --help`
Attempting with retry: make prepare
---
skip untracked path cpu-usage.csv during rustfmt invocations
skip untracked path src/doc/book/ during rustfmt invocations
skip untracked path src/doc/rust-by-example/ during rustfmt invocations
skip untracked path src/llvm-project/ during rustfmt invocations
Diff in /checkout/compiler/rustc_lint/src/levels.rs at line 187:
         intravisit::walk_field_def(self, s);
 
-    fn visit_variant(
-        &mut self,
-        &mut self,
-        v: &'tcx hir::Variant<'tcx>,
-    ) {
+    fn visit_variant(&mut self, v: &'tcx hir::Variant<'tcx>) {
         self.add_id(v.id);
         intravisit::walk_variant(self, v);
     }
Running `"/checkout/obj/build/x86_64-unknown-linux-gnu/stage0/bin/rustfmt" "--config-path" "/checkout" "--edition" "2021" "--unstable-features" "--skip-children" "--check" "/checkout/compiler/rustc_lint/src/nonstandard_style.rs" "/checkout/compiler/rustc_lint/src/levels.rs" "/checkout/compiler/rustc_lint/src/types.rs" "/checkout/compiler/rustc_lint/src/enum_intrinsics_non_enums.rs" "/checkout/compiler/rustc_lint/src/context.rs" "/checkout/compiler/rustc_lint/src/hidden_unicode_codepoints.rs" "/checkout/compiler/rustc_lint/src/late.rs" "/checkout/compiler/rustc_mir_build/src/thir/cx/block.rs"` failed.
If you're running `tidy`, try again with `--bless`. Or, if you just want to format code, run `./x.py fmt` instead.

[bors]

☔ The latest upstream changes (presumably #101064) made this pull request unmergeable. Please resolve the merge conflicts.

[fee1-dead]

Closing in favor of #101620. Thank you for working on this @cjgillot!