Update cc-rs to 1.2.39

[dpaoliello]

For my purposes, contains fixes when compiling the Rust compiler for Arm64EC.

Checked the commits since 1.2.16, and I don't see anything else that may affect Rust?

find-msvc-tools was also factored out from cc to allow updating the use in rustc_codegen_ssa (finding the linker when running the Rust compiler) to be separate from the use in rustc_llvm (building LLVM as part of the Rust compiler).

[rustbot]

Failed to set assignee to NobodyXu: invalid assignee

Note: Only org members with at least the repository "read" role, users with write permissions, or people who have commented on the PR may be assigned.

[rustbot]

These commits modify the Cargo.lock file. Unintentional changes to Cargo.lock can be introduced when switching branches and rebasing PRs.

If this was unintentional then you should revert the changes before this PR is merged.
Otherwise, you can ignore this comment.

Some changes occurred in compiler/rustc_codegen_ssa

cc @WaffleLapkin

[jieyouxu]

@bors try jobs=aarch64-msvc-1,aarch64-apple,test-various,dist-apple-various

[rust-bors]

☀️ Try build successful (CI)
Build commit: d8ac12e (d8ac12e3b0cef5271092800d62ab852417973355, parent: a1208bf765ba783ee4ebdc4c29ab0a0c215806ef)

[jieyouxu]

Let me know @dpaoliello if you want to try with find-msvc-tools, otherwise I'm fine with the current change as well.
@rustbot author

[rustbot]

Reminder, once the PR becomes ready for a review, use @rustbot ready.

[dpaoliello]

I switched to find-msvc-tools, since I was here changing things.

@rustbot ready

[rustbot]

The list of allowed third-party dependencies may have been modified! You must ensure that any new dependencies have compatible licenses before merging.

cc @davidtwco, @wesleywiser

[davidtwco]

@bors r=davidtwco,jieyouxu

[bors]

📌 Commit 45a0e10 has been approved by davidtwco,jieyouxu

It is now in the queue for this repository.

[dpaoliello]

Done

@rustbot ready

[jieyouxu]

Thanks

View changes since this review

[jieyouxu]

@bors r+ rollup=never

[bors]

📌 Commit 826c925 has been approved by jieyouxu

It is now in the queue for this repository.

[bors]

⌛ Testing commit 826c925 with merge bc1d727...

[bors]

☀️ Test successful - checks-actions
Approved by: jieyouxu
Pushing bc1d727 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 c9537a9 (parent) -> bc1d727 (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 bc1d7273dfbc6f8a11c0086fa35f6748a13e8d3c --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-loongarch64-musl: 10893.4s -> 5156.0s (-52.7%)
2. dist-loongarch64-linux: 11354.2s -> 5538.0s (-51.2%)
3. pr-check-1: 1890.9s -> 1300.6s (-31.2%)
4. aarch64-apple: 8224.5s -> 6892.9s (-16.2%)
5. armhf-gnu: 4783.2s -> 5547.6s (16.0%)
6. dist-apple-various: 3959.6s -> 4533.5s (14.5%)
7. x86_64-gnu-llvm-20-1: 3700.0s -> 3255.6s (-12.0%)
8. dist-i586-gnu-i586-i686-musl: 5332.8s -> 4750.4s (-10.9%)
9. dist-x86_64-apple: 6383.1s -> 7051.9s (10.5%)
10. x86_64-gnu-llvm-20-2: 5415.8s -> 5973.3s (10.3%)

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 (bc1d727): comparison URL.

Overall result: ❌ regressions - 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)	0.7%	[0.3%, 3.7%]	259
Regressions ❌ (secondary)	0.9%	[0.0%, 3.0%]	287
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.7%	[0.3%, 3.7%]	259

Max RSS (memory usage)

Results (primary -3.1%, secondary -3.9%)

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)	-	-	0
Improvements ✅ (primary)	-3.1%	[-3.1%, -3.1%]	1
Improvements ✅ (secondary)	-3.9%	[-4.5%, -3.2%]	3
All ❌✅ (primary)	-3.1%	[-3.1%, -3.1%]	1

Cycles

Results (primary -1.9%, secondary 2.2%)

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)	2.2%	[2.2%, 2.2%]	1
Improvements ✅ (primary)	-1.9%	[-2.5%, -1.1%]	3
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-1.9%	[-2.5%, -1.1%]	3

Binary size

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

Bootstrap: 474.143s -> 475.525s (0.29%)
Artifact size: 390.63 MiB -> 390.06 MiB (-0.14%)

[jieyouxu]

... I did not expect that, feel free to post a revert if this isn't noise, won't have chance to investigate until EOD tmrw.

[lqd]

@dpaoliello this PR wasn’t expected to impact compile-time performance this much, right?

[dpaoliello]

No, I wouldn't have thought so. I may have missed a change that would affect it.
@NobodyXu thoughts about the perf regression?

[madsmtm]

I investigated the regression a bit, I don't think it's rustc_llvm that's the actual issue, the perf regression seem to happen on check builds as well.

A hunch could be that blake3 is compiled differently, it uses cc::Build::is_flag_supported (which changed in rust-lang/cc-rs#1424) to try to determine whether to compile C or Rust code, and it might be that one is faster than the other? Though then again, blake3 isn't used much.

Alternatively it might be tikv-jemalloc-sys, the default flags emitted there might affect it.

[Kobzol]

We saw similar-ish wins (in the opposite direction) when -flto started being passed to jemalloc. Maybe that regressed now?

[Kobzol]

Reverted in #148298.

@rustbot label: +perf-regression-triaged