Dirty area is not update properly while animation with requestAnimationFrame enabled for EFL port. Visit here http://ie.microsoft.com/testdrive/Graphics/RequestAnimationFrame/Default.html.
Created attachment 112643 [details] Proposed patch
*** Bug 70900 has been marked as a duplicate of this bug. ***
Informal r+ on my side.
Passing "-b single http://ie.microsoft.com/testdrive/Graphics/RequestAnimationFrame/Default.html" to EWebLauncher did not show anything weird while rendering the linked page. Can you describe what was supposed to be seen?
(In reply to comment #4) > Passing "-b single http://ie.microsoft.com/testdrive/Graphics/RequestAnimationFrame/Default.html" to EWebLauncher did not show anything weird while rendering the linked page. Can you describe what was supposed to be seen? I guess you missed applying a submitted patch for bug67112 before. This will enable requestAnimationFrame JS API so that you can see 2 clocks and some changing numbers.
> Passing "-b single http://ie.microsoft.com/testdrive/Graphics/RequestAnimationFrame/Default.html" to EWebLauncher did not show anything weird while rendering the linked page. Can you describe what was supposed to be seen? Hi, Kubo, I think that you can notice the reason why this patch is reasonable if you check www.daum.net. Just go to www.daum.net and scroll down 2 times. and then click daum title image on top of the left cornor of website without this patch first. And finally use this patch for the same checking.
LGTM.
(In reply to comment #5) > (In reply to comment #4) > > Passing "-b single http://ie.microsoft.com/testdrive/Graphics/RequestAnimationFrame/Default.html" to EWebLauncher did not show anything weird while rendering the linked page. Can you describe what was supposed to be seen? > > I guess you missed applying a submitted patch for bug67112 before. This will enable requestAnimationFrame JS API so that you can see 2 clocks and some changing numbers. Oops, thanks for the pointer. From what I see, the patch makes callback counters actually work, right?
(In reply to comment #8) > Oops, thanks for the pointer. From what I see, the patch makes callback counters actually work, right? BTW, if that's the issue, doesn't it need to be fixed in the tiled backing store too?
(In reply to comment #9) > (In reply to comment #8) > > Oops, thanks for the pointer. From what I see, the patch makes callback counters actually work, right? > > BTW, if that's the issue, doesn't it need to be fixed in the tiled backing store too? You are right. Tiled view also requires it.
(In reply to comment #10) > (In reply to comment #9) > > (In reply to comment #8) > > > Oops, thanks for the pointer. From what I see, the patch makes callback counters actually work, right? > > > > BTW, if that's the issue, doesn't it need to be fixed in the tiled backing store too? > > You are right. > Tiled view also requires it. OK, so if this is the issue that this patch fixes, you have my informal r+, even though it'd be better to fix both backing stores in a single patch.
Comment on attachment 112643 [details] Proposed patch rubber-stamp. This really should be tested, but I don't think EFL has DumpREnderTree yet... Does it? If so, we could use repaint tests.
Comment on attachment 112643 [details] Proposed patch Clearing flags on attachment: 112643 Committed r99041: <http://trac.webkit.org/changeset/99041>
All reviewed patches have been landed. Closing bug.