[MPlayer-dev-eng] [PATCH] Direct3D VO better handling of uncooperative video adapter

[Reimar Döffinger]

On Tue, Jan 27, 2009 at 12:39:16AM +0200, Georgi Petrov wrote:
> With the current code when the adapter is uncooperative, attempts are
> being made to execute the actual D3D instructions (like StretchRect,
> Begin/End scence) and so on, which leads to a crash (even if those
> functions all return errors and then all other frame-copy functions
> return error, it's not the right thing to do anyway).

If that is correct, there is indeed a race-condition and the issues will
continue to exist completely unchanged if the adapter becomes
uncooperative e.g. in the middle of a draw_slice function.
In which case the patch does not _fix_ anything, it only reduces the
probability of problems, and how much it reduces that might well depend
on the specific graphics drivers and whether you use the OSD.
As I see it, with any driver update things might break completely again,
even with that patch applied (yes, it is very unlikely, coding based on
luck still seems wrong to me).