[PATCH v3] ws2_32: Check if waiting asyncs exist in WSARecv/WSASend

[Zebediah Figura (she/her)]

On 5/11/21 10:03 PM, Dongwan Kim wrote:
> WSARecv call should not intercept message for the waiting overlapped WSARecv.
> To do that, We should know if waiting asyncs exist before trying ws2_recv/ws2_send.
> 

Hello, thanks for the patch!

I don't think this is quite the right way to handle this 
architecturally. However, more importantly, this introduces extra 
overhead into what can already be a rather slow operation. Hence two 
questions:

(1) Is there an application that actually breaks because of this? (If 
not, I think it's probably not worth doing.)

(2) Can you write a test case—ideally to be submitted to Wine's test 
suite along with the patch, but failing that a standalone program—to 
demonstrate that this really is true?