Prefer remapping the relative library/ and compiler/ directories

[Urgau]

This is done to avoid leaking the relative paths to the standard library after the overall of filenames in #149709.

Noted that the paths were already leaking before, but to a lesser extent since most (but not all) the paths embedded in the distributed rlib were absolute.

In general Cargo compiles workspace members with relative paths, so it's better anyway to remap the relative path.

In addition to our tests I have manually confirmed that it also works as expected for the printed diagnostics paths.

cf. https://rust-lang.zulipchat.com/#narrow/channel/131828-t-compiler/topic/remapping.20of.20the.20standard.20library/near/564093571

r? @jieyouxu

[Urgau]

@bors try

[rust-bors]

☀️ Try build successful (CI)
Build commit: 7b8f45c (7b8f45c196d5f10b2f859b09c58205f9f9228066, parent: 68f11a11b666b36137374b6e54fc0d013a3b26cf)

[jieyouxu]

Thanks, let give this a try at least. If this proves problematic, we can revert.

View changes since this review

[jieyouxu]

@bors r+ rollup=never

[bors]

📌 Commit 74eecf9 has been approved by jieyouxu

It is now in the queue for this repository.

[bors]

⌛ Testing commit 74eecf9 with merge 0e7ebb4...

[bors]

💔 Test failed - checks-actions

[jieyouxu]

Huh.

[Urgau]

Those tests failures are interesting, they show the remapped paths, despite compiletest asking rustc to change the prefix to /rustc/FAKE_PREFIX (with -Zsimulate-remapped-rust-src-base=/rustc/FAKE_PREFIX).

I have manually checked and the above prefix change works as expected for imported files (ie files described in rmeta), but as far as I can tell none of those filenames comes from an rlib/rmeta.

Instead they come from the compiled artifact it-self, and more specifically the debuginfo, which is retrived by the backtraces and shown on those tests. (yes, even in the note:, thanks to some wild backtrace capturing in rustc_errors)

The reason this wasn't an issue before is because were only producing the relative paths, which are not remapped before this PR. As for why those tests didn't fail on PR CI, I have no idea.

---

In any-case we well need a mechanism that supplants the current /rustc/FAKE_PREFIX normalization.

This is even complicated by the fact depending on whenever the compiler/std was compiled with remapping on or off changes the kind of paths we get, as with remapping on we will get absolute paths (starting with /rustc/) but with remapping off we will only get relative paths, which I think prevents us from applying any normalization globally.

Given that all of those test already had custom normalization for those backtrace paths, I have opted to add a simple //@ normalize-stderr: "/rustc(?:-dev)?/[a-z0-9.]+/" -> "" rule to all of the failing tests.

[Urgau]

@bors try jobs=dist-x86_64-linux

[rust-bors]

☀️ Try build successful (CI)
Build commit: 3822c58 (3822c58804c3d0f64dd4e612587894ac319d5d22, parent: fcf67da039f42e3905cf6f69e33304299c45149f)

[jieyouxu]

Hm... yeah. The extra test normalizations don't feel great, but not end of the world.

View changes since this review

[jieyouxu]

@bors r+

[bors]

📌 Commit 0acc846 has been approved by jieyouxu

It is now in the queue for this repository.

[bors]

⌛ Testing commit 0acc846 with merge 1d8f9c5...

[bors]

☀️ Test successful - checks-actions
Approved by: jieyouxu
Pushing 1d8f9c5 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 0f0d850 (parent) -> 1d8f9c5 (this PR)

Test differences

No test diffs found

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard 1d8f9c548d4d8845951f1d927caac1620e86353d --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. dist-apple-various: 3558.8s -> 4164.5s (+17.0%)
2. x86_64-gnu-miri: 4470.2s -> 5032.7s (+12.6%)
3. x86_64-mingw-2: 9121.5s -> 10068.3s (+10.4%)
4. x86_64-gnu-llvm-20-2: 5548.6s -> 6084.8s (+9.7%)
5. aarch64-msvc-2: 6565.1s -> 5978.0s (-8.9%)
6. aarch64-apple: 10310.5s -> 9417.7s (-8.7%)
7. tidy: 148.8s -> 161.5s (+8.5%)
8. x86_64-msvc-2: 8305.3s -> 8956.8s (+7.8%)
9. dist-aarch64-apple: 8173.8s -> 7545.4s (-7.7%)
10. dist-i586-gnu-i586-i686-musl: 5510.0s -> 5103.6s (-7.4%)

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 (1d8f9c5): 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%]	3
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (primary 0.6%, secondary 0.8%)

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

	mean	range	count
Regressions ❌ (primary)	2.9%	[1.5%, 4.4%]	2
Regressions ❌ (secondary)	2.5%	[2.5%, 2.5%]	1
Improvements ✅ (primary)	-4.2%	[-4.2%, -4.2%]	1
Improvements ✅ (secondary)	-0.9%	[-0.9%, -0.9%]	1
All ❌✅ (primary)	0.6%	[-4.2%, 4.4%]	3

Cycles

Results (primary 2.1%, secondary 2.8%)

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

	mean	range	count
Regressions ❌ (primary)	2.1%	[2.1%, 2.2%]	2
Regressions ❌ (secondary)	2.8%	[2.2%, 4.1%]	5
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	2.1%	[2.1%, 2.2%]	2

Binary size

Results (primary 0.1%, secondary 0.3%)

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%]	4
Regressions ❌ (secondary)	0.3%	[0.2%, 0.5%]	5
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.1%	[0.1%, 0.1%]	4

Bootstrap: 479.055s -> 480.26s (0.25%)
Artifact size: 390.50 MiB -> 390.57 MiB (0.02%)