Don't install msys2 in CI

[ChrisDenton]

windows-msvc doesn't need it and windows-gnu installs its own version

try-job: dist-x86_64-msvc
try-job: dist-i686-msvc
try-job: dist-aarch64-msvc
try-job: dist-i686-mingw
try-job: dist-x86_64-mingw
try-job: dist-x86_64-msvc-alt

[ChrisDenton]

@bors try

[bors]

⌛ Trying commit c94ab20 with merge 6896d6e...

[bors]

☀️ Try build successful - checks-actions
Build commit: 6896d6e (6896d6e9adfe1dd2de222e2a597a970ef8064fc0)

[ChrisDenton]

It seems like this is now entirely unnecessary

r? infra-ci

[ChrisDenton]

Wait let me just try the dist jobs, just to be sure nothing is using it.

@bors try

[bors]

⌛ Trying commit c94ab20 with merge 92de1572c8bf8fb3f3cf1d8493c666346368ca36...

[mati865]

windows-gnu installs its own version

It has nothing to do with MSYS2. The linked code installs mingw-w64 GNU toolchain.

[ChrisDenton]

Right I should have said "its own tools". Things like a bash emulator are already installed on the host and we don't need to install a package manager for anything.

[Kobzol]

Hmm, I remember several attempts to fix this and it always failed, would be cool if we can indeed remove it :D We still run a makefile in Windows CI, I'm kind of surprised that it runs without msys2. Can Windows just execute the Makefile now?

[ChrisDenton]

I believe the last Makefile has now just been removed in #135572 (though maybe after the try run here). But yes I did expect that to fail. Surprising it didn't but in any case that shouldn't be an issue going forward.

[Kobzol]

I didn't mean run-make tests, but src/bootstrap/mk/Makefile.in, which is used to run CI tests on Windows.

[ChrisDenton]

Oh, huh. Maybe the newer runners have make installed?

I guess my next question would be, do we need make for running Windows tests in CI?

[Kobzol]

I don't think so, it's mostly just a historical artifact. I'll try to get rid of it.

[bors]

☀️ Try build successful - checks-actions
Build commit: 92de157 (92de1572c8bf8fb3f3cf1d8493c666346368ca36)

[ChrisDenton]

Those succeeded too so should this be blocked on removing the testing Makefile or not?

[Kobzol]

We don't need to block on it. Since all these jobs passed, probably Makefile is installed somehow by default. Let's try it!

@bors r+ rollup=never

(Marking as rollup=never since perhaps some other jobs might be affected, who knows)

[bors]

📌 Commit c94ab20 has been approved by Kobzol

It is now in the queue for this repository.

[mati865]

Oh, huh. Maybe the newer runners have make installed?

I guess my next question would be, do we need make for running Windows tests in CI?

IIRC there is MSYS2 preinstalled in GHA images for some time now, so make is still available.

[matthiaskrgr]

@bors p=4

[bors]

⌛ Testing commit c94ab20 with merge e060723...

[bors]

☀️ Test successful - checks-actions
Approved by: Kobzol
Pushing e060723 to master...

[rust-timer]

Finished benchmarking commit (e060723): 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
Regressions ❌ (secondary)	0.2%	[0.2%, 0.2%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (primary 2.8%, secondary 2.4%)

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)	2.8%	[2.8%, 2.8%]	1
Regressions ❌ (secondary)	2.4%	[2.4%, 2.4%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	2.8%	[2.8%, 2.8%]	1

Cycles

Results (secondary -0.7%)

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)	2.6%	[2.6%, 2.6%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-4.0%	[-4.0%, -4.0%]	1
All ❌✅ (primary)	-	-	0

Binary size

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

Bootstrap: 778.956s -> 778.909s (-0.01%)
Artifact size: 329.06 MiB -> 329.06 MiB (0.00%)