https://bugs.winehq.org/show_bug.cgi?id=46099
--- Comment #34 from Zebediah Figura z.figura12@gmail.com --- Created attachment 62897 --> https://bugs.winehq.org/attachment.cgi?id=62897 ntdll: Add a futex-based implementation of WaitOnAddress().
Thanks for the confirmation.
I have another patch here, which implements WaitOnAddress() on top of futexes. It's a bit of an unorthodox approach, but is the best I was able to come up with. I'd appreciate it if people could test it (applied on top of the other patches in this thread) with a real application.
Fabian, this patch fails your tests, because it treats WakeByAddressSingle() like WakeByAddressΑll() and wakes up every thread. Since spurious wakeups should always be anticipated by the application, I doubt this should break anything, but since you wrote the tests, are you aware of any application that depends on that behaviour?