<rdar://76837320> and not only on the root of the move.
Created attachment 426955 [details] Patch
Comment on attachment 426955 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=426955&action=review > Source/WebCore/ChangeLog:9 > + While moving a subtre, we invalidate the floating object list so that we don't end up with incorrectly placed floats (they'll get regenerated during the subsequent layout). We should really stop moving subtrees. Also spelling.
(In reply to Antti Koivisto from comment #2) > Comment on attachment 426955 [details] > Patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=426955&action=review > > > Source/WebCore/ChangeLog:9 > > + While moving a subtre, we invalidate the floating object list so that we don't end up with incorrectly placed floats (they'll get regenerated during the subsequent layout). > > We should really stop moving subtrees. Also spelling. Yes and yes. I wonder what we'll come up with for LFC where the layout nodes (renderers) don't store layout states so moving subtrees is nothing but a simple re-parenting activity, while we have to run invalidation on the formatting state objects.
Created attachment 426979 [details] [fast-cq] Patch
Committed r276549 (236989@main): <https://commits.webkit.org/236989@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 426979 [details].