[2/2] kernel32/tests: Add a test to show that WaitCommEvent should fail if there is a pending wait.

[Dmitry Timoshkov]

Marvin <testbot at winehq.org> wrote:

> === w7pro64 (32 bit comm) ===
> comm.c:850: Test failed: WaitCommEvent failed with a timeout
> comm.c:862: recovering after WAIT_TIMEOUT...
> comm.c:874: Test failed: WaitCommEvent error 0
> comm.c:876: Test failed: WaitCommEvent: expected EV_TXEMPTY, got 0
> comm.c:880: WaitCommEvent for EV_TXEMPTY took 1014 ms (timeout 1000)
> comm.c:882: Test failed: WaitCommEvent used 1014 ms for waiting
> comm.c:970: recovering after WAIT_TIMEOUT...
> comm.c:1549: test_AbortWaitCts timeout 500 handle 00000130
> comm.c:1516:  Changing CommMask on the fly for handle 00000130 after timeout 500

These are all existing tests, nothing to do with this patch. Also it's
strange that 1/2 has passed these tests without failures while 2/2 fails.

This new testbot behaviour doesn't look very robust.

-- 
Dmitry.