GraphicsContext3D should allow its hostWindow to be NULL
Created attachment 104045 [details] Patch
Comment on attachment 104045 [details] Patch Attachment 104045 [details] did not pass chromium-ews (chromium-xvfb): Output: http://queues.webkit.org/results/9401279
Created attachment 104050 [details] Patch Patch leveldatabase DEPS as in Chrome
Comment on attachment 104050 [details] Patch Attachment 104050 [details] did not pass chromium-ews (chromium-xvfb): Output: http://queues.webkit.org/results/9407269
The cr-linux EWS bot compiled successfully the first time, but failed a number of international font layout tests (not sure why). On subsequent runs, it failed to build since the leveldatabase directory became unversioned, presumably because other jobs were run in between. As for the layout test failures, I'm not sure what's going on, since the WebKit canaries are running with ToT and seem fine with it. I think the change is correct, but I'll keep an eye on both the canary and webkit.org bots on landing.
Comment on attachment 104050 [details] Patch Attachment 104050 [details] did not pass cr-mac-ews (chromium): Output: http://queues.webkit.org/results/9405252
Comment on attachment 104050 [details] Patch R=me. Please watch the bots carefully when landing.
Committed r93131: <http://trac.webkit.org/changeset/93131>
Created attachment 104138 [details] fixes clipping weirdness
(In reply to comment #9) > Created an attachment (id=104138) [details] > fixes clipping weirdness Did I mess something up? Or is this uploaded to the wrong bug?
Gah fail, I mismerged ChangeLogs and webkit-patch sent this to the wrong bug :(. Sorry...