Also, you did verify on Windows that interlaced and dmDisplayFixedOutput are in ascending order, right?
No I didn't, this is only to match `winemac.drv` logic and simplify the lookup there. I could perhaps find a monitor with interlaced modes but I don't think I have anything that would have the stretched thing.
In any case it's only there to keep the current behavior the same, and only `winemac.drv` uses these flags. If the behavior is wrong, it may need to be changed but I don't think it's the purpose of this MR to do that and it could be done later.
So a/b_stretched will either be TRUE or FALSE. But there are DMDFO_DEFAULT, DMDFO_CENTER, and DMDFO_STRETCH. That doesn't seem right.
In the same way, only `winemac.drv` uses these flags, and it's either DMDFO_CENTER, or DMDFO_STRETCH. If some driver needs the full set of flags at some point it'll need to be changed but right now this is only meant to move current logic, not change it.
I prefer these codes to be moved right before their use so we can save a few CPU cycles if the check can return early. The same can be said for width and height. It probably doesn't matter with modern compiler optimization, but still...
Sure.