tytung, is this anything useful?
0001-Forward-port-surface-dithering-from-CM10.2.patchersinozler wrote:1. Black screen issue is spreading (Google Search + Currents App + Facebook App) (Workaround: disable apps' Hardware Acceleration via Hwa Settings - Not Helpfull)
The thing is: All these Apps. works fine on their welcome screen, but when you choice another section or make a search or select something, that new screen comes black but navigation buttons is visible and working.
As I mentioned the other day, my theory is that it's related to Android's HTML rendering components. The screen goes black in Google Now/Search only when you do a search that requires web results. My guess is that the Currents and Facebook apps are querying the same HTML rendering component of Android. It makes it easier on the app developers, after all, to use web data whenever possible (especially when all of Facebook is on the web already, anyways), instead of needlessly duplicating everything in the app. If we can determine what that HTML rendering component is and either fix it or replace it, we'll likely fix all those apps that depend on it. That's my theory, at least. I believe that I read something about Google replacing the old WebView renderer with a Chromium-based one in KitKat (or something like that), so that could be the underlying change. Beyond that, I don't really know enough to pursue the problem solving much further, though. Hopefully, tytung can use this info to find the solution.
EDIT: To add to this, through some searching, I found that this is/was a problem on KitKat ROMs for other old phones, like the Desire HD (similar to the HD2) and the Nexus S. In a Nexus S thread, I found this guy's post:
http://forum.xda-developers.com/showpos ... tcount=366That's what I was thinking, too, that the web renderer is making a call to graphics libraries that's failing. That would be why you get only a black square, yet you can click around on it to access the links that you can't see (meaning that the actual data is there).
Also, he followed up with this large post that suggests that he made some progress:
http://forum.xda-developers.com/showpos ... tcount=382...then this one that suggests that the CM team fixed the bug:
http://forum.xda-developers.com/showpos ... tcount=393I wonder why we're still seeing it, then. Perhaps the changes aren't in the particular versions used in our ROMs or they're enough to fix the issue for that chipset, but not for ours. I don't know, but I'm throwing everything out there in hopes that tytung knows what (if anything) to do with it.