fix close_read_wakes_up test

[lukas-code]

On windows, shutdown does not interrupt read, even though we document that it does (see #121594).

The close_read_wakes_up test has a race condition and only passes on windows if the shutdown happens before the read. This PR ignores the test on windows adds a sleep to make it more likely that the read happens before the shutdown and the test actually tests what it is supposed to test on other platforms.

I'm submitting this before any docs changes, so that we can find out on what platforms shutdown actually works as documented.

r? @ChrisDenton

[ChrisDenton]

I do strongly disagree with my past self for enabling this test. As I recall at the time, there were a number of tests that were simply disabled on Windows instead of fixing the bug that caused the failure and when bugs were fixed, tests weren't re-enabled. It seems I was a bit, let's say "over enthusiastic", about re-enabling tests.

In any case, let's see what CI has to say.

@bors r+ rollup=iffy

[bors]

📌 Commit 9abe47e has been approved by ChrisDenton

It is now in the queue for this repository.

[bors]

⌛ Testing commit 9abe47e with merge 48a15aa...

[bors]

☀️ Test successful - checks-actions
Approved by: ChrisDenton
Pushing 48a15aa to master...

[rust-timer]

Finished benchmarking commit (48a15aa): comparison URL.

Overall result: ❌ regressions - no action needed

@rustbot label: -perf-regression

Instruction count

This is a highly reliable metric that was used to determine the overall result at the top of this comment.

	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

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)	-3.8%	[-5.3%, -2.2%]	2
All ❌✅ (primary)	-	-	0

Cycles

Results

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)	-2.2%	[-2.2%, -2.2%]	1
All ❌✅ (primary)	-	-	0

Binary size

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

Bootstrap: 650.027s -> 650.646s (0.10%)
Artifact size: 172.57 MiB -> 172.57 MiB (0.00%)