Replying back to the list. Based on your description, well, yes, it could be.
It also could be that someone at Samsung decided to optimize away some synchronization primitives that normally protect Skia calls. But... personally I don't have any knowledge of the internals of Android graphics, and certainly don't have the sources for Samsung's Note firmware... so it's just pure speculation on my part, and as such, it's probably gotten long past having any meaning... -- K PS As far as moderation goes - a few days for new subscribers, I believe. Also note that messages appear to come in irregular batches. 2012/5/2 Martin Heller <[email protected]> > Hi, > > my suspicion is that it might have something to do with the Skia > Graphics Engine and its thread safety. > > This is what I have posted on the mailing list but it hasn't got through > the moderation yet: > > After looking at the stack trace and seeing two threads in > android.graphics.Canvas.native_drawBitmap(Native Method) > > I looked into Android source code, discovered that Skia engine is used > and a quick Google search 'skia thread safe' eventually lead me to > > > http://groups.google.com/group/android-developers/browse_thread/thread/3db72fe9b3481b53 > > Might this be related? > > Martin > > P.S. Do you know how the moderation works and how long it should take? > > > On 05/02/2012 09:45 AM, Kostya Vasilyev wrote: > > > > 02.05.2012 4:50, abe oppenheim написал: > >> if this is the case, any idea why: > >> > >> 1. it isn't fixed by adding Thread.sleep(100) to the background thread? > > > > Why would it? > > > > The issue might be caused by concurrent data modification, so adding > > sleep lowers the probability, but does not really change anything. There > > is probably still a missing synchronization primitive somewhere. > > > >> 2. it only occurs on the one device? > > > > More likely it's one device *model* rather than one *device*. > > > > Now as to why, that's a philosophical question, really. Firmware bugs > > and weirdnesses happen, but Samsung devices seem to be special :) > > > > In any case, if you're going to / did contact Samsung, a stack trace > > will make your bug report more complete. > > > > I also understand that the Note will soon be officially updated to > > Android 4.0 (although Samsung typically doesn't update all countries at > > once, like HTC or SE... which makes it less useful from a developer's > > point of view). > > > > -- K > > > >> > >> On Tue, May 1, 2012 at 8:17 PM, Streets Of Boston > >> <[email protected] <mailto:[email protected]>> wrote: > >> > >> Yep, both threads are in Canvas.native_drawBitmap and clobbering > >> each other (or least the background thread is clobbering the UI > one). > >> This is a tough one to fix or to work-around. > >> > >> > >> On Tuesday, May 1, 2012 7:27:00 PM UTC-4, Ab wrote: > >> > >> I hadn't, but I just did on your suggestion. > >> > >> The call chain I see in eclipse looks the same as one in the > >> traces.txt file: > >> > >> http://pastebin.com/hummn87H > >> > >> It appears that the thread gets wedged while drawing one of > >> the ImageButtons. > >> > >> On Tuesday, 1 May 2012 19:09:19 UTC-4, Kostya Vasilyev wrote: > >> > >> Abe, > >> > >> Have you tried pausing the UI thread in the debugger once > >> it's locked up, in order to inspect its call chain? > >> > > > > -- > > Kostya Vasilyev > > > > -- > > You received this message because you are subscribed to the Google > > Groups "Android Developers" group. > > To post to this group, send email to [email protected] > > To unsubscribe from this group, send email to > > [email protected] > > For more options, visit this group at > > http://groups.google.com/group/android-developers?hl=en > -- You received this message because you are subscribed to the Google Groups "Android Developers" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/android-developers?hl=en
