Printing seems to use software snapshotting, and is for some reason flipped (meaning I must have missed a hisorical step in r198242's changelog. Here are the failing tests: Regressions: Unexpected image-only failures (3) printing/quirks-percentage-height-body.html [ ImageOnlyFailure ] printing/quirks-percentage-height.html [ ImageOnlyFailure ] printing/standards-percentage-heights.html [ ImageOnlyFailure ]
When these tests were added (r141459), we were in the same state we are in today (no flipping in WKTR). Something must have regressed elsewhere in the meantime, or something.
http://trac.webkit.org/changeset/190306 looks highly suspicious to me.
> When these tests were added (r141459), we were in the same state we are in today (no flipping in WKTR). Something must have regressed elsewhere in the meantime, or something. ACTUALLY, when these tests were added they were immediately skipped in Mac WebKit2 and then not unskipped until r190306 (by which time the WKTR flipping logic had been broken, and they were only passing because of that).
I have temporarily skipped these tests in <http://trac.webkit.org/projects/webkit/changeset/198285>
Created attachment 274217 [details] Patch
Comment on attachment 274217 [details] Patch Clearing flags on attachment: 274217 Committed r198302: <http://trac.webkit.org/changeset/198302>
All reviewed patches have been landed. Closing bug.