Regression Test - kernel atom test

[Alexandre Julliard]

"Dmitry Timoshkov" <dmitry at baikal.ru> writes:

> Probably NT4 just doesn't set last error here. But we have to check for
> last error values, because it will be very hard to track down this problem
> if some application chokes on it. What do we have to do? Drop the test for
> last error? That's not an option IMO.

Such an application would die on NT4 too, so it's unlikely to
exist. And yes, if the error codes are not set consistently under
Windows then there is no point in testing them, we might as well drop
the test.

> This should be investigated. Win2000 apparently has more strict tests
> for parameter validation and I don't see a reason why we shouldn't do
> the same.

Because it hurts performance, and any app that depends on the strict
validation will die on NT4 anyway. Plus it will hide potential bugs in
our code.

> I have a suggestion then. How about removing conflicting tests but make
> Wine behave as win2000 (and actually as described in MSDN)?

There is no reason to try to be more compatible than Windows
itself. It just creates extra work and doesn't improve our
compatibility at all. Our time is much better spent fixing things
where we actually need to be compatible.

-- 
Alexandre Julliard
julliard at winehq.com