https://reviews.llvm.org/D71440

I rewrote the test case source file for TestExprDoesntBlock.py so that it uses std::thread rather than pthreads. So that test case should at least compile on Windows, and so I prospectively removed the Windows xfail from the test. But I don’t actually know how well suspending & resuming threads works on Windows. I’ll keep an eye on the bots but if somebody sees this fail again, please just put the expectedFail back. This wouldn’t be a new failure caused by this patch, just previously untried code that still fails…

Thanks,

Jim