Fix test compiling Fuchsia with rust-under-test by erickt · Pull Request #162573 · rust-lang/rust

erickt · 2026-09-10T03:57:31Z

The rust tests for Fuchsia were disabled some time ago in #138690 when fuchsia.googlesource.com was temporarily rate limiting git checkout requests during a high period of requests. This makes a number of attempts to avoid this issue:

It checks out Fuchsia with git filter tree:0 to reduce download size.
It blocks downloading Fuchsia's Rust toolchain.
It adds a number of retries to requests.
It reduces the parallel git checkout threads.

Since it has been a while since we updated the Fuchsia integration, it also updates:

Use the most recent commit be532c.
It updates to the clang commit 5e63f2.
It updates to the Fuchsia SDK version 33.20260909.5.1.
It updates the ubuntu container to 26.04.
It renames the test job to test-x86_64-fuchsia.
The Fuchsia build system now requires shared libraries to be stripped.
The Fuchsia build now requires the Rust toolchain to have the artifacts for arm64 and riscv64 to be present, even though they might not be used for this build.

Note that I did use Gemini to help debug some issues, but all the code has been written by me.

try-job: dist-various-2
try-job: test-x86_64-fuchsia

rustbot · 2026-09-10T03:57:35Z

Warning

If you are changing how CI LLVM is built or linked, make sure to bump
src/bootstrap/download-ci-llvm-stamp.

cc @jieyouxu

rustbot · 2026-09-10T03:57:37Z

r? @marcoieni

rustbot has assigned @marcoieni.
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

Why was this reviewer chosen?

The reviewer was selected based on:

Owners of files modified in this PR: infra-ci
infra-ci expanded to Kobzol, Mark-Simulacrum, jieyouxu, marcoieni, ubiratansoares
Random selection from Mark-Simulacrum, jieyouxu, marcoieni, ubiratansoares

erickt · 2026-09-10T04:15:03Z

It looks like the job-linux-8c config was removed in #161562. For now I switched to using job-linux-4c to at least verify the tests are still working, but should I instead switch over to job-linux-x86-8c-ec2? Also, I saw in #162063 that the aarch64 bots might be about half as much as the x64 bots, should we switch to targeting them instead? We test that configuration in our Fuchsia tests, so it should work out of the box.

erickt · 2026-09-10T04:16:34Z

@bors try

Fix test compiling Fuchsia with rust-under-test try-job: dist-various-2 try-job: x86_64-fuchsia

rust-bors · 2026-09-10T04:17:31Z

💔 Test for a03d4d5 failed: CI. Failed job:

Calculate job matrix (web logs, enhanced plaintext logs)

rustbot · 2026-09-10T04:26:15Z

The rustc-dev-guide subtree was changed. If your future PRs only touch the subtree, consider submitting them directly to rust-lang/rustc-dev-guide, which is where the document is primarily maintained (and has faster CI).

cc @BoxyUwU, @tshepang

erickt · 2026-09-10T04:27:11Z

@bors try

Fix test compiling Fuchsia with rust-under-test try-job: dist-various-2 try-job: test-x86_64-fuchsia

rust-bors · 2026-09-10T04:32:25Z

💔 Test for c09146d failed: CI. Failed jobs:

try - dist-various-2 (web logs, enhanced plaintext logs)
try - test-x86_64-fuchsia (web logs, enhanced plaintext logs)

erickt · 2026-09-10T04:37:40Z

Looks like I need to incorporate some changes from CI, so I switched this to a draft mode until I get the tests passing.

@bors try

erickt · 2026-09-16T20:20:57Z

@bors try

Fix test compiling Fuchsia with rust-under-test try-job: dist-various-2 try-job: test-x86_64-fuchsia

rust-bors · 2026-09-16T21:33:12Z

☀️ Try build successful (CI)
Build commit: 005a85e (005a85e0293b203a5d777a70526ded250896be32)
Base parent: 92801ed (92801ede88f6c10e34df6976a4cfde40142585f2)

jieyouxu

Thanks
@bors r+ rollup

View changes since this review

rust-bors · 2026-09-17T12:03:29Z

📌 Commit 8ed9008 has been approved by jieyouxu

It is now in the queue for this repository.

jieyouxu · 2026-09-17T12:03:47Z

@bors rollup=never

Fix test compiling Fuchsia with rust-under-test The rust tests for Fuchsia were disabled some time ago in #138690 when fuchsia.googlesource.com was temporarily rate limiting git checkout requests during a high period of requests. This makes a number of attempts to avoid this issue: * It checks out Fuchsia with git filter `tree:0` to reduce download size. * It blocks downloading Fuchsia's Rust toolchain. * It adds a number of retries to requests. * It reduces the parallel git checkout threads. Since it has been a while since we updated the Fuchsia integration, it also updates: * Use the most recent commit [be532c]. * It updates to the clang commit [5e63f2]. * It updates to the Fuchsia SDK version 33.20260909.5.1. * It updates the ubuntu container to 26.04. * It renames the test job to `test-x86_64-fuchsia`. * The Fuchsia build system now requires shared libraries to be stripped. * The Fuchsia build now requires the Rust toolchain to have the artifacts for arm64 and riscv64 to be present, even though they might not be used for this build. Note that I did use Gemini to help debug some issues, but all the code has been written by me. try-job: dist-various-2 try-job: test-x86_64-fuchsia [be532c]: https://fuchsia.googlesource.com/integration/+/be532c365aa6ba56b08ed684c0f8bb71960c18fd [5e63f2]: llvm/llvm-project@5e63f2c

rust-bors · 2026-09-18T09:52:44Z

💔 Test for b763e7e failed: CI. Failed job:

auto - test-x86_64-gnu-distcheck (web logs, enhanced plaintext logs)

JonathanBrouwer · 2026-09-18T10:02:42Z

@bors retry

rust-bors · 2026-09-18T14:13:22Z

☀️ Test successful - CI
Approved by: jieyouxu
Duration: 3h 1m 35s
Pushing 420ed2a to main...

github-actions · 2026-09-18T14:16:10Z

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 498b89d (parent) -> 420ed2a (this PR)

Test differences

Show 1 test diff

1 doctest diff were found. These are ignored, as they are noisy.

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard 420ed2a0c3d7225b1744266fd884d431b4d8cfe0 --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

dist-i686-linux: 1h 14m -> 1h 54m (+54.8%)
dist-x86_64-llvm-mingw: 1h 31m -> 2h 12m (+45.0%)
test-x86_64-gnu-stdlib-semver-check: 17m 42s -> 9m 59s (-43.6%)
test-x86_64-gnu-llvm-22-2: 1h 12m -> 1h 41m (+39.2%)
dist-x86_64-musl: 1h 43m -> 2h 22m (+38.0%)
test-x86_64-gnu-llvm-22-1: 1h 20m -> 50m 39s (-37.2%)
test-x86_64-msvc-ext3: 2h 2m -> 1h 20m (-33.9%)
dist-x86_64-illumos: 1h 29m -> 1h 59m (+32.8%)
test-x86_64-gnu-distcheck: 1h 48m -> 2h 21m (+30.7%)
dist-armhf-linux: 1h 33m -> 1h 5m (-29.9%)

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 · 2026-09-18T15:33:09Z

Finished benchmarking commit (420ed2a): comparison URL.

Overall result: no relevant changes - no action needed

@rustbot label: -perf-regression

Instruction count

This perf run didn't have relevant results for this metric.

Max RSS (memory usage)

Results (primary -5.2%)

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)	-	-	0
Improvements ✅ (primary)	-5.2%	[-5.2%, -5.2%]	1
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-5.2%	[-5.2%, -5.2%]	1

Cycles

Results (secondary -1.0%)

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)	3.1%	[3.1%, 3.1%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-3.0%	[-3.4%, -2.6%]	2
All ❌✅ (primary)	-	-	0