Pretend in bootstrap snapshot tests that we always build in-tree LLVM

[jieyouxu]

Otherwise, depending on whether CI LLVM is inhibited or if an externally-provided LLVM is used, bootstrap host LLVM build step could be missing in step snapshots.

Note that I'm not sure if this is the right solution (this might be a solution). I imagine we do want to control for the set of configuration that these snapshot tests are run, as much as possible.

r? @Kobzol

[jieyouxu]

I'm not 100%, would like to double-check this change in #143126.

[dianqk]

IIUC, this is because the CI configuration is changing the test input?

[Kobzol]

We will need to normalize more things in the bootstrap test config, yeah. I don't understand why the LLVM build step disappeared though, when LLVM was modified, it should be built, not downloaded, so if anything the step should appear (but really it should be called in both cases).

[jieyouxu]

Yeah, it should be unconditionally called (a no-op step is fine), but it should still appear...

[Kobzol]

@bors r+ p=1 (bootstrap fix blocking other PRs)

[bors]

📌 Commit 006e97a has been approved by Kobzol

It is now in the queue for this repository.

[bors]

⌛ Testing commit 006e97a with merge 6dec76f...

[bors]

☀️ Test successful - checks-actions
Approved by: Kobzol
Pushing 6dec76f 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 fd9ca71 (parent) -> 6dec76f (this PR)

Test differences

No test diffs found

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard 6dec76f1c2809fded082dd44d3752d3f6220d767 --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: 3402.2s -> 5001.2s (47.0%)
2. dist-apple-various: 7127.3s -> 5850.2s (-17.9%)
3. x86_64-apple-1: 7105.8s -> 8287.6s (16.6%)
4. aarch64-apple: 4334.0s -> 3679.7s (-15.1%)
5. tidy: 70.7s -> 78.8s (11.4%)
6. test-various: 4584.4s -> 5068.9s (10.6%)
7. dist-aarch64-msvc: 8031.4s -> 8868.5s (10.4%)
8. dist-x86_64-apple: 8971.3s -> 9667.6s (7.8%)
9. i686-gnu-nopt-2: 7560.9s -> 8131.9s (7.6%)
10. x86_64-gnu-aux: 5968.5s -> 6397.0s (7.2%)

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 (6dec76f): 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)	-	-	0
Improvements ✅ (secondary)	-0.0%	[-0.0%, -0.0%]	1
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

This benchmark run did not return any relevant results for this metric.

Cycles

Results (secondary -0.9%)

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)	4.4%	[4.1%, 4.7%]	2
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-4.5%	[-6.9%, -3.2%]	3
All ❌✅ (primary)	-	-	0

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 458.797s -> 461.008s (0.48%)
Artifact size: 372.11 MiB -> 372.14 MiB (0.01%)