Re: [PATCH v25 0/4] MR3504: ntdll: An implementation of SRWLOCK that closer matches Windows'.

[Zebediah Figura (＠zfigura)]

Zebediah Figura (@zfigura) commented about dlls/kernel32/tests/sync.c:
> -
>    /* function is not yet in XP, only in newer Windows */
> -
>    win_skip("no srw lock support.\n");
> -
>    return;
> - }
> -
> - *u.l = 0;
> - pReleaseSRWLockExclusive(&srwlock_example);
> - ok(*u.l == 0xffffffff, "expected 0xffffffff, got %lx\n", *u.l);
> -
> - *u.l = 1;
> - pReleaseSRWLockExclusive(&srwlock_example);
> - ok(*u.l == 0, "expected 0x0, got %lx\n", *u.l);
> -
> - *u.l = 0x10000;
> - pReleaseSRWLockExclusive(&srwlock_example);
> - ok(*u.l == 0xffff, "expected 0xffff, got %lx\n", *u.l);

It turns out the first and third tests actually crash for me on Windows 7 with STATUS_RESOURCE_NOT_OWNED, and abort without even triggering an exception on Windows 10. I don't know if the testbot complains, but even if it doesn't, that might be because of Windows 10's weird failure mode.