TaskScheduler test is flaky on Windows

[weliveindetail]

Test Case 'TaskSchedulerTests.testHighPriorityTasksGetExecutedBeforeLowPriorityTasks' started at 2025-02-27 04:31:50.411

C:\Users\swift-ci\jenkins\workspace\swift-PR-windows\sourcekit-lsp\Tests\SemanticIndexTests\TaskSchedulerTests.swift:46: error: TaskSchedulerTests.testHighPriorityTasksGetExecutedBeforeLowPriorityTasks : XCTAssertTrue failed - [Set([high(0)]), Set([high(1), high(0)]), Set([high(1), high(2), high(0)]), Set([high(2), high(1), high(0), high(3)]), Set([high(2), high(1), high(3)]), Set([high(2), high(3)]), Set([high(3)]), Set([]), Set([high(4)]), Set([high(4), high(6)]), Set([high(4), high(6), high(5)]), Set([high(7), high(4), high(6), high(5)]), Set([high(7), high(6), high(5)]), Set([high(7), high(6)]), Set([high(6)]), Set([]), Set([high(8)]), Set([high(9), high(8)]), Set([low(0), high(9), high(8)]), Set([low(1), low(0), high(9), high(8)]), Set([low(1), high(9), high(8)]), Set([high(9), high(8)]), Set([high(8)])] did not fulfill predicate

Test Case 'TaskSchedulerTests.testHighPriorityTasksGetExecutedBeforeLowPriorityTasks' failed (1.219 seconds)

This causes some noise in pre-merge testing:
https://ci-external.swift.org/job/swift-PR-windows/37033/ 27 Feb, PR #67986
https://ci-external.swift.org/job/swift-PR-windows/36986/ 26 Feb, PR #79628
https://ci-external.swift.org/job/swift-PR-windows/36987/ 26 Feb, PR #79624

The underlying issue is that Swift.org bots for the mainline Windows toolchain are lacking coverage and developers have no access to the configuration. LLDB tests are one example for this (see swiftlang/swift#79020). They run in PR-testing, but not in mainline testing and thus we have no good monitoring for flakes.

[weliveindetail]

There was a recent commit actually, that might have fixed it a32be44

[ahoppen]

Synced to Apple’s issue tracker as rdar://145750493

[ahoppen]

Yes, #2012 addressed that flakiness.