http://bugs.winehq.org/show_bug.cgi?id=2398
------- Additional Comments From mail@science.su 2007-28-03 14:26 -------
My impression is that submitting a patch to revert back to X child windows won't be accepted, especially if it introduces new problems.
"Proper and correct solution will not be accepted, let's use hacks and tricks!". I'm sorry, but I really don't understand this point of view. I can understand this in case if we already have working hack with no regressions and problems but we havn't and unlikely to have one. At least this is my impression and also general opinion: hacks almost always introduce additional problems or simply don't work in all cases and currently this is exactly the case with Ulrich patches.
So why don't accept this patch? Personally I use it for very long time (since it creation) already, run games, applications, professional software and see no problems at all (except need to keep it in sync with current git and maybe a little dirty code but this unrelated and possible to fix). No performance problems, no regressions - what else is needed? If I (or anyone else) miss any problems related to using X child windows it would be good idea to post these problem here, but my opinion for now that this patch works very well and currently is only real solution.