tools-aux ci runner: also cross-test doctests in Miri

[RalfJung]

Miri now supports running doctests across different targets. Let's use that to run the std doctests on aarch64-apple-darwin, i686-pc-windows-msvc.

try-job: x86_64-gnu-aux

[rustbot]

r? @albertlarsan68

rustbot has assigned @albertlarsan68.
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

[RalfJung]

I don't know why I have split up this call into separate --no-doc and --doc invocations... we'll see if anything breaks I guess.^^

[RalfJung]

I think I may have done it to get separate timings. It doesn't seem worth keeping that.

[RalfJung]

@bors try

[bors]

⌛ Trying commit 73fcdf6 with merge 1d4757c...

[RalfJung]

@bors try

[bors]

⌛ Trying commit 721fb52 with merge 7b5d67f...

[bors]

💔 Test failed - checks-actions

[RalfJung]

@bors try

[bors]

⌛ Trying commit 6eb1afa with merge 1777446...

[bors]

☀️ Try build successful - checks-actions
Build commit: 1777446 (177744690ff9c236ed94d4de85d63183f702c3ed)

[RalfJung]

While we're at it I also enabled strict provenance enforcement for some of the tests -- we probably don't want to have int2ptr casts in the standard library. (miri-test-libstd already tests with that flag.)

@bors try

[bors]

⌛ Trying commit 8b7b026 with merge 3449d7b...

[bors]

💔 Test failed - checks-actions

[RalfJung]

@bors try

[bors]

⌛ Trying commit b9e6995 with merge 8418ba9...

[bors]

☀️ Try build successful - checks-actions
Build commit: 8418ba9 (8418ba98707c96d9d8c43d30d3d5a4b3e5ea5088)

[RalfJung]

Oh, the labels are outdated.
@rustbot ready

[RalfJung]

@rust-lang/miri this only takes a few extra minutes since doctests are executed in parallel by rustdoc -- so maybe we can land this without waiting for an infra review?

[saethlin]

I was thinking of approving this based on your description, but I'm surprised to see edits to this file in particular. I've never touched it before, and I don't know how it integrates with our CI. So I'm not confident this is the right way to implement the change you want.

[RalfJung]

That's the file that controls which tests are run on the aux builder. It is the only file where we could possibly implement this change.

[saethlin]

Oh I see, there are two kind of jobs, those that set SCRIPT and those that use this mechanism.

I'll approve this since it's just following the existing pattern but I feel like at some point we were trying to get away from Makefiles?

@bors r+

[bors]

📌 Commit b9e6995 has been approved by saethlin

It is now in the queue for this repository.

[RalfJung]

Yeah this makefile feels like a relict from the past. But I don't have it in me to understand enough of our CI pipeline to clean this up.

[workingjubilee]

@bors rollup=iffy

[bors]

⌛ Testing commit b9e6995 with merge b17dba4...

[bors]

☀️ Test successful - checks-actions
Approved by: saethlin
Pushing b17dba4 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 99426c5 (parent) -> b17dba4 (this PR)

Test differences

Show 2 test diffs

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

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard b17dba45186c454576d0fc8fb93ecc65eb1a763a --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: 4457.5s -> 5130.5s (15.1%)
2. dist-x86_64-apple: 8301.8s -> 9431.4s (13.6%)
3. dist-aarch64-apple: 6723.0s -> 5834.0s (-13.2%)
4. x86_64-apple-1: 7034.1s -> 7636.5s (8.6%)
5. dist-ohos-armv7: 4421.5s -> 4180.7s (-5.4%)
6. mingw-check-2: 2194.5s -> 2080.6s (-5.2%)
7. dist-x86_64-netbsd: 5082.5s -> 5336.3s (5.0%)
8. aarch64-apple: 5345.4s -> 5094.9s (-4.7%)
9. x86_64-gnu-llvm-19-1: 3740.4s -> 3893.0s (4.1%)
10. dist-aarch64-msvc: 8268.6s -> 8540.3s (3.3%)

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 (b17dba4): comparison URL.

Overall result: ❌ regressions - no action needed

@rustbot label: -perf-regression

Instruction count

This is the most reliable metric that we have; it was used to determine the overall result at the top of this comment. However, even this metric can sometimes exhibit noise.

	mean	range	count
Regressions ❌ (primary)	0.4%	[0.4%, 0.4%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.4%	[0.4%, 0.4%]	1

Max RSS (memory usage)

Results (secondary -2.6%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.6%	[-2.6%, -2.6%]	1
All ❌✅ (primary)	-	-	0

Cycles

Results (secondary 3.2%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	3.2%	[3.2%, 3.2%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Binary size

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

Bootstrap: 744.362s -> 743.703s (-0.09%)
Artifact size: 372.29 MiB -> 372.32 MiB (0.01%)