[Webkit-unassigned] [Bug 106884] [EFL] Coordinated Graphics Deadlock when running fast/block/abspos-child-container-changes-from-relative-to-static.html

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=106884


Dominik Röttsches (drott) <dominik.rottsches at intel.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[EFL] NWRT stalling with    |[EFL] Coordinated Graphics
                   |repaint timeouts after ~8   |Deadlock when running
                   |seconds                     |fast/block/abspos-child-con
                   |                            |tainer-changes-from-relativ
                   |                            |e-to-static.html
                 CC|                            |mikhail.pozdnyakov at intel.co
                   |                            |m, noam at webkit.org




--- Comment #4 from Dominik Röttsches (drott) <dominik.rottsches at intel.com>  2013-01-15 05:52:53 PST ---
When running fast/block/abspos-child-container-changes-from-relative-to-static.html WebKitTestRunner forces a repaint on the WebProcess but we end up in a deadlock and WTR detects a repaint failure in dumpPixelsAndCompareWithExpected.

As far as I can see it, this happens because, first
CoordinatedLayerTreeHost::forceRepaintAsync is called,
which then calls scheduleLayerFlush() - after this timer fires:
performScheduledLayerFlush() is called, but here:
m_waitingForUIProcess is true and the method returns early, while the UIProcess is waiting for the WebProcess.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.