Re: [PATCH 0/1] MR5258: d3d9/tests: Skip desktop window tests if device creation fails.

[Zebediah Figura (＠zfigura)]

So normally I object to this kind of patch in the context of "fixing" Wine tests, since it doesn't really fix anything—a failure is a failure. This is obviously a bit of a different situation though, but I guess the point is that this is something that I'd sign off on only for dxvk's benefit, not because it'd be desirable "anyway".

I'm still curious what the end goal will look like here—this patch makes a lot more sense if we also have "todo_dxvk" annotations in tree everywhere—but if it helps for now I'm not opposed to accepting anyway, at least until we work that out.