Stop collecting unmentioned constants

[Mark-Simulacrum]

This avoids generating useless dead LLVM IR. This appears to have regressed and/or been introduced in #53821 (unfortunately a very large PR - I don't see any direct discussion there of this particular change), but as far as I can tell is at least no longer necessary -- or we lack test coverage -- because none of our UI tests indicate diagnostics regressions. The adjusted codegen-units test has comments explicitly noting that these items should not be collected ("These are not referenced, so they do not produce mono-items").

I noticed this while looking at libcore LLVM IR we generate, which contained dead code references to the NOOP Waker item, which is never used inside libcore. Producing LLVM IR for it during libcore's compilation, only for that IR to get deleted by LLVM as unused, isn't useful. Note that the IR is generally all marked internal, too.

[Mark-Simulacrum]

@bors2 try @rust-timer queue

[rust-bors]

⌛ Trying commit 928564c with merge 499813f…

To cancel the try build, run the command @bors2 try cancel.

[rust-bors]

☀️ Try build successful (CI)
Build commit: 499813f (499813f1e930d596afe1c2fc1ec25fcff26a6473, parent: 111e9bc64bbdce14122e3676978f2ceefa5bff1a)

[rust-timer]

Finished benchmarking commit (499813f): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

Benchmarking this pull request means it may be perf-sensitive – we'll automatically label it not fit for rolling up. You can override this, but we strongly advise not to, due to possible changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please do so in sufficient writing along with @rustbot label: +perf-regression-triaged. If not, please fix the regressions and do another perf run. If its results are neutral or positive, the label will be automatically removed.

@bors rollup=never
@rustbot label: -S-waiting-on-perf +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)	13.0%	[0.7%, 21.2%]	3
Regressions ❌ (secondary)	9.0%	[5.9%, 13.4%]	5
Improvements ✅ (primary)	-1.5%	[-2.2%, -0.8%]	9
Improvements ✅ (secondary)	-34.3%	[-67.9%, -0.2%]	4
All ❌✅ (primary)	2.1%	[-2.2%, 21.2%]	12

Max RSS (memory usage)

Results (primary -5.6%, secondary -0.5%)

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

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	3.5%	[3.5%, 3.5%]	1
Improvements ✅ (primary)	-5.6%	[-5.6%, -5.6%]	1
Improvements ✅ (secondary)	-4.5%	[-4.5%, -4.5%]	1
All ❌✅ (primary)	-5.6%	[-5.6%, -5.6%]	1

Cycles

Results (primary 3.3%, secondary -11.0%)

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

	mean	range	count
Regressions ❌ (primary)	15.0%	[12.8%, 17.1%]	2
Regressions ❌ (secondary)	8.1%	[5.0%, 12.0%]	5
Improvements ✅ (primary)	-2.5%	[-2.7%, -2.1%]	4
Improvements ✅ (secondary)	-58.8%	[-58.8%, -58.8%]	2
All ❌✅ (primary)	3.3%	[-2.7%, 17.1%]	6

Binary size

Results (primary -0.1%, secondary 12.3%)

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

	mean	range	count
Regressions ❌ (primary)	0.1%	[0.1%, 0.1%]	1
Regressions ❌ (secondary)	12.3%	[11.7%, 14.0%]	4
Improvements ✅ (primary)	-0.1%	[-0.2%, -0.1%]	9
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-0.1%	[-0.2%, 0.1%]	10

Bootstrap: 689.093s -> 689.162s (0.01%)
Artifact size: 371.98 MiB -> 371.87 MiB (-0.03%)

[Mark-Simulacrum]

The regressions look likely to be noise - specifically codegen unit shuffling - (I'll look in more detail later) but overall this seems like what we want, ctfe-stress improvement in particular looks exactly like what we want to see here.

Marking as waiting on review.

[rustbot]

r? @oli-obk

rustbot has assigned @oli-obk.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

[oli-obk]

but as far as I can tell is at least no longer necessary -- or we lack test coverage -- because none of our UI tests indicate diagnostics regressions.

we walk all free constants and evaluate them, so we won't miss them this way (as we want to evaluate even unused free constants). But we definitely don't need to collect and codegen them

@bors r+

[bors]

📌 Commit 928564c has been approved by oli-obk

It is now in the queue for this repository.

[workingjubilee]

Everyone takes another turn

@bors p=1

[bors]

⌛ Testing commit 928564c with merge 13c46fd...

[bors]

☀️ Test successful - checks-actions
Approved by: oli-obk
Pushing 13c46fd to master...

[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 d51b6f9 (parent) -> 13c46fd (this PR)

Test differences

No test diffs found

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard 13c46fd0b089360922a557d8e18a63a2c41dfbeb --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. x86_64-apple-2: 4098.2s -> 6082.6s (48.4%)
2. dist-apple-various: 7470.2s -> 8759.0s (17.3%)
3. dist-x86_64-apple: 8441.0s -> 9693.8s (14.8%)
4. dist-x86_64-netbsd: 4609.1s -> 5188.0s (12.6%)
5. dist-ohos-x86_64: 4022.2s -> 4387.9s (9.1%)
6. aarch64-apple: 4603.9s -> 4955.7s (7.6%)
7. x86_64-apple-1: 8062.9s -> 8638.6s (7.1%)
8. dist-aarch64-apple: 5986.6s -> 5571.8s (-6.9%)
9. dist-i586-gnu-i586-i686-musl: 5106.7s -> 5453.2s (6.8%)
10. x86_64-msvc-ext3: 5881.1s -> 6253.0s (6.3%)

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 (13c46fd): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

Our benchmarks found a performance regression caused by this PR.
This might be an actual regression, but it can also be just noise.

Next Steps:

• If the regression was expected or you think it can be justified,
  please write a comment with sufficient written justification, and add
  @rustbot label: +perf-regression-triaged to it, to mark the regression as triaged.
• If you think that you know of a way to resolve the regression, try to create
  a new PR with a fix for the regression.
• If you do not understand the regression or you think that it is just noise,
  you can ask the @rust-lang/wg-compiler-performance working group for help (members of this group
  were already notified of this PR).

@rustbot label: +perf-regression
cc @rust-lang/wg-compiler-performance

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)	19.2%	[17.2%, 21.2%]	2
Regressions ❌ (secondary)	9.1%	[6.0%, 13.4%]	5
Improvements ✅ (primary)	-1.6%	[-2.3%, -0.8%]	8
Improvements ✅ (secondary)	-67.8%	[-67.9%, -67.8%]	2
All ❌✅ (primary)	2.6%	[-2.3%, 21.2%]	10

Max RSS (memory usage)

Results (primary -1.7%, secondary 1.3%)

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

	mean	range	count
Regressions ❌ (primary)	4.0%	[4.0%, 4.0%]	1
Regressions ❌ (secondary)	4.9%	[2.4%, 9.4%]	3
Improvements ✅ (primary)	-4.5%	[-5.0%, -4.0%]	2
Improvements ✅ (secondary)	-4.0%	[-4.1%, -3.9%]	2
All ❌✅ (primary)	-1.7%	[-5.0%, 4.0%]	3

Cycles

Results (primary 2.4%, secondary -8.8%)

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

	mean	range	count
Regressions ❌ (primary)	14.9%	[13.0%, 16.9%]	2
Regressions ❌ (secondary)	7.2%	[2.5%, 13.2%]	6
Improvements ✅ (primary)	-2.7%	[-3.1%, -2.0%]	5
Improvements ✅ (secondary)	-41.0%	[-59.0%, -4.9%]	3
All ❌✅ (primary)	2.4%	[-3.1%, 16.9%]	7

Binary size

Results (primary -0.2%, secondary 12.3%)

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

	mean	range	count
Regressions ❌ (primary)	0.1%	[0.1%, 0.1%]	1
Regressions ❌ (secondary)	12.3%	[11.7%, 14.0%]	4
Improvements ✅ (primary)	-0.2%	[-0.3%, -0.1%]	15
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-0.2%	[-0.3%, 0.1%]	16

Bootstrap: 692.293s -> 690.533s (-0.25%)
Artifact size: 372.06 MiB -> 372.10 MiB (0.01%)